Full text of "Poly(amide-graft-acrylate) interfacial compounds"

POLY ( AMI DE-GRAFT-ACRYLATE) INTERFACIAL COMPOUNDS

By
MICHAEL PEREZ ZAMORA

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT

OF THE REQUIREMENTS FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1997

This dissertation is dedicated to all those who supported me

emotionally, physically, and financially throughout my

seemingly endless years of education; especially my wife

Karen, my parents Pablo and Sylvia, and my grandmother

Bella; and to the loving memory of Oscar Zamora and Manuel

Leon .

ACKNOWLEDGMENTS

I cannot express enough gratitude and appreciation to
my advisor and doctoral committee chairman, Dr. Anthony
Brennan, for his years of teaching, guidance and support in
areas that extended well beyond my academic endeavors. I
would also like to express sincere thanks to the members of
my supervisory committee for their advice and teaching: Dr.
Jim Adair, Dr. Chris Batich, Dr. Eugene Goldberg, and Dr.
Ken Wagener.

I consider this experience substantially enriched by
the support of my colleagues. Whether there was
collaboration, debate or simply encouragement, it would have
been endlessly more difficult without the following: Dr.
Michael Antonell, Dr. Drew Amery, Craig Habeger, Jeff
Kerchner, Dr. James Marotta, Jeremy Mehlem, Dr. Rodrigo
Orefice, Luxsamee Plangsmangas, Mark Schwarz, and Dr. Chris
Widenhouse .

Special thanks are extended to Arthur Gavrin for his
assistance with NMR spectroscopy and for the lively
discussions in his short time here, and Ananth Naman and Dr.
Rob Chodelka for their unwavering support and friendship. I
would be remiss not to mention Jesse Arnold and Tom Miller,
my colleagues of five years with whom I have grown, matured,

iii

and learned. I will forever be appreciative of our
continuous discussions, debates, and collaboration as well
as their undying support.

:i v

TABLE OF CONTENTS

ACKNOWLEDGMENTS iii

LIST OF TABLES viii

LIST OF FIGURES ix

ABSTRACT xiii

1 . INTRODUCTION 1

1.1. Poly (amide-g-acrylate) Graft Copolymers 1

1.1.1. Macromonomers 1

1.1.2. Addition-Condensation Graft Copolymers 3

1.1.3. Applications 4

1.2. Offsetting Polymerization Shrinkage in Dental
Resins through the Incorporation of Maleic
Anhydride 5

2 . BACKGROUND 7

2.1. Graft Copolymers 8

2.1.1. Synthetic Routes to Graft Copolymers 8

2.1.1.1. Anionically polymerized macromonomers 9

2.1.1.2. Macromonomers through chain transfer

f unctionalization 9

2.1.1.3. Modeling chain transfer reaction 15

2.1.2. Polyamide Addition-Condensation Multiphase
Copolymers 19

2.1.2.1. Poly (amide-b-olef in) block copolymers.... 19

2.1.2.2. Poly (amide-g-olef in) graft copolymers.... 28

2.2. Polymerization Shrinkage in Dental Composites 35

2.2.1. History of Dental Composites 36

2.2.2. Reduction of Polymerization Shrinkage in
Dental Resins 39

2.2.3. Use of Anhydrides in Dental Applications 41

2.2.4. Filler Modification in the Reduction of
Overall Composite Shrinkage 42

3 . MATERIALS AND METHODS 4 4

3.1. Materials 44

3.1.1. Macromonomer Reactants 44

3.1.2. Graft Copolymer Reactants 45

3.1.3. Dental Monomers 45

3.2. Methods 46

3.2.1. Synthesis and Characterization of
Macromonomers 4 6

3.2.2. Synthesis and Characterization of Graft
Copolymers 4 9

3.2.3. Synthesis and Characterization of Anhydride
Modified Dental Resins 52

RESULTS AND DISCUSSION 55

4.1. Synthesis and Characterization of Amino Acid-
terminated Poly (acrylate) Macromonomers using

Chain Transfer Chemistry 55

4.1.1. Determination of an Appropriate Solvent

System 59

4.1.2. Preliminary Studies of the Effectiveness of
Cysteine as a Chain Transfer Agent in the
Polymerization of Butyl Acrylate 61

4.1.3. Determination of Chain Transfer Constant of
Cysteine in the Synthesis of Amino Acid-
terminated Poly (butyl acrylate) Macromonomers 75

4.1.4. Cysteine Chain Transfer in the Synthesis of
Amino Acid-terminated Poly (methyl
methacrylate : octaf luoropentyl methacrylate)
Macromonomers 8 8

4.2. Polyamide-g-poly (acrylate) graft copolymers

from Amino Acid-terminated Macromonomers 99

4.2.1. Synthesis of Poly (amide-g-butyl acrylate) .... 99

4.2.2. Characterization of Graft Copolymers 104

4.2.3. Blends of Graft Copolymers with Nylon 6 139

4.3. Offsetting Polymerization Shrinkage in Dental
Resins through the Incorporation of Maleic
Anhydride 144

4.3.1. Copolymer Compositions 146

4.3.2. Copolymer Characterization 149

SUMMARY AND CONCLUSIONS 158

5.1. Chain Transfer Functionalization of

Poly (acrylates) and Poly (methacrylates) 161

5.1.1. Conclusions for Preliminary Evaluation of

Cysteine Chain Transfer Agent 161

5.1.2 Conclusions for Synthesis and
Characterization of Amino Acid-terminated

Poly (butyl acrylate) 162

5.1.3 Conclusions for Synthesis and
Characterization of Amino Acid-terminated

Poly (MMA-co-OFPMA) 162

5.2. Graft Copolymerizations of Macromonomers with

Polyamide Precursors 163

5.2.1. Conclusions for Synthesis of Poly (amide-g-

acrylate) Graft Copolymers 164

5.2.2. Conclusions for Characterization of

Poly (amide-g-acrylate) Graft Copolymers 165

5.2.3. Conclusions for Mechanical Properties of

Nylon 6/Graf t Copolymer Blends 166

5.3. Offsetting Polymerization Shrinkage in

Poly (dimethacrylate) Dental Resins 166

5.3.1. Conclusions for Characterization of Maleic
Anhydride-containing Dental Resins 167

5.3.2. Conclusions for Synthesis and
Characterization of Anhydride Copolymer with

PEMA 168

6. FUTURE WORK 169

6.1. Macromonomers and Graft Copolymers 169

6.1.1. Macromonomer Work 169

6.1.2. Graft Copolymers 170

6.2. Anhydride-containing Dental Resins 171

LIST OF REFERENCES 173

BIOGRAPHICAL SKETCH 181

Vll

LIST OF TABLES
TABLE page

4.1 Solvent determination for monomer and chain

transfer agent °0

4.2 Percent f unctionalization versus molar mass for

poly (butyl acrylate) macromonomers 87

4.3 Molar mass values of f luoroacrylate copolymers from
GPC 91

4.4 Percent weight loss from Soxhlet extraction for
polyamide graft copolymers 105

4.5 Chemical composition of purified graft copolymers from
elemental analysis and NMR 137

4.6 Inherent viscosities of polyamide graft copolymers. 138

4.7 Tensile properties of nylon 6 blends 140

4.8 Experimental matrix of dental monomer compositions. 148

4.9 PEMA-maleic anhydride monomer compositions 14 9

4.10 EWC of maleic anhydride dental resins 150

4.11 Residual weight gain, anhydride incorporation and post
polymerization expansion of maleic anhydride dental
resins 151

4.12 Glass transition temperatures and composition of PEMA-
maleic anhydride copolymers 155

4.13 Molar mass averages from GPC for PEMA-anhydride
copolymers 157

LIST OF FIGURES
Figure page

1.1 Schematic illustration of general graft copolymer
structure 1

2.1. Mechanism of f unctionalization using chain transfer
agents 11

2.2. Schematic illustration of utilized chain transfer
agents and macromonomers thereof 16

2.3. Mechanism of block copolymer formation through
sequential polymerizations of vinyl monomer and
isocyanates 21

2.4. Reaction schematic for macroinitiator formation and
subsequent anionic block polymerization of caprolactam.

2.5 Reaction schematic of block copolymerization initiated
by nitrosated polyamide macroinitiators 25

2 . 6 Reaction schematic of AIBN containing polyamide and
block copolymer thereof 27

2.7 Reaction schematic of in situ graft copolymer formation
from glycidyl methacrylate copolymers 31

2.8 Reaction schematic of in situ graft copolymer formation
from maleic anhydride modified polyolefins copolymers..

2.9 Mechanism of ring opening polymerization of spiro
orthocarbonate monomers 40

4.1. Mechanism of functionalization using chain transfer

reactions 57

4.2 Schematic of ideal amino acid functionalization during
polymerization of butyl acrylate 62

1 X

4.3 GPC results of preliminary butyl acrylate

polymerizations 64

4 . 4 DSC trace of side product 66

4.5 FTIR spectra of side product of cysteine modified

P (BA) 67

4.6 Comparison of FTIR spectra of side product with butyl
acrylate monomer 68

4.7 Reaction pathway of cysteine with acrylates 69

4.8 Structure and elemental analysis of side product. ... 70

4.9 Side reaction preventing complete chain transfer. ... 73

4.10 Effect of acidification on chain transfer
functionalization of poly (butyl acrylate) 76

4.11 GPC results of polymerizations of butyl acrylate
varying cysteine concentrations 80

4.12 Mayo plot for chain transfer constant determination for
cysteine rbutyl acrylate system 81

4.13 FTIR spectra of poly (butyl acrylate) and cysteine end-
capped p(BA) macromonomer . Macromonomer is the

1.2 kg/mo 1 p(BA) synthesized using 1000:64:1 butyl
acrylate : cysteine :AIBN mole ratio 83

4.14 1 H-NMR spectra of neat poly (butyl acrylate) 84

4.15 H-NMR spectra of 1.2 kg/mo 1 cysteine modified
poly(butyl acrylate) 85

4.16 13 C-NMR spectra of 1.2kg/mol cysteine modified

poly (butyl acrylate) 86

4.17 Chemical structure of f luoroacrylate copolymer 90

4.18 GPC result of chain transfer polymerizations of

f luoroacrylate copolymers 91

4.19 Comparison of effectiveness of chain transfer for p(BA)
and f luoroacrylate copolymer 93

4.20 x H-NMR spectra of MMA-OFPMA macromonomer 95

4.21 Mayo plot for chain transfer constant determination for
cysteine :MMA-co-OFPMA system 96

4.22 FTIR spectra of MMA-OFPMA copolymers 98

4.23 Amide-acrylate graft copolymer structure 100

4.24 Triphenyl phosphite driven amide formation 102

4.25 Synthesized graft copolymer compositions 103

4.26 Molar mass distributions of dissolved polymer in THF
extractant solution of 33PABA-g-66BA (refractive index
detector) 107

4.27 UV spectra of dissolved polymer in THF extractant
solution of 33PABA-g-66BA from GPC-UV 108

4.28 Molar mass distributions of dissolved polymer in THF
extractant solution of 33PhDAA-g-66BA (refractive index
detector) 110

4.29 UV spectra of dissolved polymer in THF extractant
solution of 33PhDAA-g-66BA from GPC-UV 112

4.30 Molar mass distributions of dissolved polymer
in THF extractant solution of 10PhDAA-g-90BA
(refractive index detector) 113

4.31 UV spectra of dissolved polymer in THF extractant
solution of 10PhDAA-g-90BA from GPC-UV 114

4.32 Molar mass distributions of dissolved polymer in THF
extractant solution of 33PhDAA-g-66FA (refractive index
detector) 116

4.33 UV spectra of dissolved polymer in THF extractant
solution of 33PhDAA-g-66FA from GPC-UV 117

4.34 Molar mass distributions of dissolved polymer in THF
extractant solution of 33PhDAA-g-66UBA (refractive
index detector) 119

4.35 UV spectra of dissolved polymer in THF extractant
solution of 33PhDAA-g-66UBA from GPC-UV 120

4.36 Transmission FTIR spectra of 66PABA-g-33BAx graft
copolymer 122

4.37 Transmission FTIR spectra of 33PABA-g-66BAx graft
copolymer 123

4.38 Transmission FTIR spectra of 33PhDAA-g-66BAx graft
copolymer 125

4.39 Transmission FTIR spectra of 10PhDAA-g-90BAx graft
copolymer 126

4.40 Transmission FTIR spectra of poly (butyl acrylate)
grafted polyamides 128

4.41 Transmission FTIR spectra of 10PhDAA-g-90BAx graft
copolymer 129

4.42 ^-NMR spectra of 66PABA-g-33BAx 131

4.43 ^-NMR spectra of 33PABA-g-66BAx 132

4.44 ^-NMR spectra of PhDAA homopolyamide 133

4.45 X H-NMR spectra of 33PhDAA-g-66BAx 134

4.46 ^-NMR spectra of 10PhDAA-g-90BAx 135

4.47 ^-NMR spectra of 33PhDAA-g-66FAx 136

4.48 TG/DTA analysis of 33PhDAA-g-66BA graft copolymer and
PhDAA homopolyamide 142

4.49 DSC analysis of 33PhDAA-g-66BA graft copolymer and
PhDAA homopolyamide 143

4.50 Chemical structures of methacrylate and anhydride
monomers for dental applications 147

4.51 TG/DTA of anhydride modified dental resin 153

4.52 FTIR spectra of poly(PEMA) and poly ( 60PEMA-co-40maleic
anhydride) 154

Xll

Abstract of Dissertation Presented to the Graduate School

of the University of Florida in Partial Fulfillment of the

Requirements for the Degree of Doctor of Philosophy

POLY(AMIDE-GRAFT-ACRYLATE) INTERFACIAL COMPOUNDS

Michael Perez Zamora

December, 1997

Chairman: Dr. Anthony B. Brennan

Major Department: Materials Science and Engineering

Graft copolymers with segments of dissimilar
chemistries have been shown to be useful in a variety of
applications as surfactants, compatibilizers, impact
modifiers, and surface modifiers. The most common route to
well defined graft copolymers is through the use of
macromonomers, polymers containing a reactive functionality
and thus capable of further polymerization. However, the
majority of the studies thus far have focused on the
synthesis of macromonomers capable of reacting with vinyl
monomers to form graft copolymers .

This study focused on the synthesis of macromonomers
capable of participating in condensation polymerizations. A
chain transfer f unctionalization method was utilized.
Cysteine was evaluated as a chain transfer agent for the
synthesis of amino acid f unctionalized poly (acrylate) and
poly (methacrylate) macromonomers. Low molar mass,
functionalized macromonomers were produced. These
macromonomers were proven to be capable of reacting with

amide precursors to form poly (amide-g-acrylate) graft
copolymers .

Macromonomers and graft copolymers were characterized
by gel permeation chromatography (GPC) , Fourier transform
infrared spectroscopy (FTIR), nuclear magnetic resonance
(NMR) spectroscopy, elemental analysis (EA) , inductively
coupled plasma (ICP), and differential scanning calorimetry
(DSC) .

The second part of this research involved
poly (dimethacrylate) dental restorative materials.
Volumetric shrinkage during the cure of these resins results
in a poor interface between the resin and the remaining
tooth structure, limiting the lifetime of these materials.
Cyclic anhydrides were incorporated into common monomer
compositions used in dental applications. Volume expansion
from the ring opening hydrolysis of these anhydrides was
shown to be feasible.

The modified dental resins were characterized by
swelling, extraction and ultraviolet spectroscopy (UV) , and
density measurements. Linear polymers designed to model the
crosslinked dental resins were characterized by FTIR, GPC,
and DSC.

CHAPTER 1
INTRODUCTION

1.1. Poly (amide-g-acrylate) Graft Copolymers

Graft copolymers are macromolecules composed of
chemically dissimilar segments in a branched architecture
(Figure 1.1) . They have been studied and utilized in a
variety of applications because of their ability to combine
the properties of their individual segments.

Figure 1.1 Schematic illustration of general graft
copolymer structure.

1.1.1 .Macromonomers

The most prevalent synthetic route to well defined
graft copolymers is through the use of macromonomers, low

molar mass polymers containing a polymerizable end group. A
review of the literature has shown that current studies,
including those within this laboratory, have concentrated on
synthesizing macromonomers which contain a residual
unsaturation at one end. Therefore, they are generally
polymerizable only with addition type monomers through a
free radical mechanism. Because the macromonomers
themselves are generally synthesized through either anionic
or free radical polymerization in the presence of a
f unctionalizing agent, the resulting graft copolymers which
can be synthesized through this method are generally limited
to addition-addition graft copolymers .

It was the objective of the first part of this study to
synthesize condensation polymerizable macromonomers,
specifically amino acid terminated macromonomers, capable of
reacting with amino acids in the synthesis of polyamide
graft copolymers. An amino acid functionality is preferred
over other end groups such as diacids or diamines due to the
inherent stoichiometry that it provides. This stoichiometry
is required in the condensation graft reaction to insure the
highest degree of polymerization possible.

The approach taken in this study involves the free
radical polymerization of acrylate and methacrylate monomers
in the presence of a functional chain transfer agent.
Mercaptans, compounds containing a sulfur-hydrogen bond, are
commonly used in chain transfer reactions. In fact,
mercaptans are commonly used to control molecular weight in

commercial polymerization reactors. Cysteine, a naturally
occurring amino acid, contains the sulfhydryl group required
for mercaptan chain transfer reactions. The addition of
cysteine, if effective as a chain transfer agent, would
result in an amino acid functionality.

1 . 1 . 2 .Addition-Condensation Graft Copolymers

There are few reports of well defined polyamide graft
copolymers with addition polymers such as poly (acrylates ) or
poly (methacrylates) . Most studies of amide graft
copolymerizations with addition polymers involve either the
in situ formation of graft copolymers in polymer blends or
radiation induced surface graft techniques. Although
effective for their intended applications, neither method
produces a well defined graft copolymer that can be isolated
and studied.

The objective of the second part of this research was
to evaluate the ability of the previously synthesized
macromonomers to participate in a condensation
polymerization of polyamide precursors. If successful, the
resulting structure would be a poly (amide-g-acrylate) graft
copolymer .

The properties of the synthesized macromonomers and
graft copolymers were characterized by a variety of
techniques including GPC, FTIR spectroscopy, NMR

spectroscopy, UV spectroscopy, elemental analysis, ICP, and
DSC.

1.1.3 .Applications

One of the most common applications of graft copolymers
is their use as compatibilizers . The appropriate graft
copolymer has been shown to migrate to the interface between
two dissimilar materials, reducing interfacial tension and
increasing the bonding at the interface.

The initial motivation for the study of polyamide-
polyacrylate graft copolymers came from the field of
dentistry. The new class of dental restorative composites,
better known as 'fillings' consist of a crosslinked
dimethacrylate matrix surrounding glass particles. The
failure of dental restorative composites, and poor
performance as compared to amalgam restorations, is
generally attributed to a poor interface. Although the
glass-resin interface has been studied extensively, the
source of failure is usually the interface between the
composite restoration and the remaining tooth structure.

Two of the main sources of this poor interface are

1) poor bonding between the exposed tooth structure,
composed of hydrophilic proteinaceous dentin tubules
and the hydrophobic dimethacrylate composite and

2) the polymerization shrinkage during composite cure
causing the restoration to pull away from the remaining

tooth structure. This leads to marginal leakage, the
infiltration of saliva and bacteria under the
restoration, which can lead to the secondary caries.
The macromonomer and graft copolymer work was targeted to
determine the ability to synthesize a copolymer capable of
interacting with both the hydrophobic methacrylate and
hydrophilic dentin structure at the tooth-restoration
interface. Although the synthesized amide-acrylate graft
copolymers were not tested in such a system and their
aromatic structures may not make them suitable for this
application, the feasibility of synthesizing the desired
structures has been evaluated.

1.2. Offsetting Polymerization Shrinkage in Dental Resins
through the Incorporation of Maleic Anhydride

The last part of this research addresses the
polymerization shrinkage problem in these dental resins.
This shrinkage is inherent to the free radical
polymerization of the multifunctional methacrylate resins
used in the dental composites. The volumetric shrinkage is
due to the reduction in molar volume, or spacing between
monomer units, that occurs when vinyl compounds are
polymerized. Polymerization shrinkage creates both a weak
interface between the tooth structure and the restoration as
well as residual stresses within the composite structure,
leading to premature failure in the restoration.

There are many research programs focused on different
chemical structures and processes that will reduce
polymerization shrinkage. Although various methods and
different monomers have been studied to alleviate this
problem, no solution to date has been discovered that
eliminates shrinkage without significantly altering the cure
and mechanical properties of the resin.

The goal of this study was to demonstrate that we can
offset the polymerization shrinkage of common dimethacrylate
resins without significantly changing the comonomer
structures through the copolymerization with maleic
anhydride. When maleic anhydride (MA) is ring opened by
hydrolysis to maleic acid, there is a corresponding
theoretical 10% increase in molar volume.

The properties of the anhydride-methacrylate
copolymers, including the ability of the anhydride to offset
the polymerization shrinkage, were characterized by
swelling, extraction and ultraviolet spectroscopy (UV) , and
density measurements, FTIR, GPC, and DSC.

CHAPTER 2
BACKGROUND

The first section of this chapter addresses how our
study of amino acid-terminated poly (acrylate) macromonomers
and poly (amide-g-acrylate) graft copolymers fits within the
massive amount of literature on the synthesis and
characterization of multiphase copolymers. An overview of
graft copolymers is given, followed by a description of the
methodology and limitations of common synthetic routes as
they relate to synthesizing addition-condensation graft
copolymers. This is followed by an analysis of other
studies concentrating specifically on polyamide graft and
block copolymers.

As was mentioned previously, the initial motivation for
amide-acrylate graft copolymers comes from the field of
dental materials. Specifically, the failure of dental
restorative materials at the tooth-resin interface is of
concern. It was mentioned that one of the main causes of
this failure is the cure shrinkage of these resins during
application, causing the resin to pull away from the tooth
surface. The second part of this chapter addresses the
problem of polymerization shrinkage in dental resins, and

analyzes the approaches described in the literature to
alleviate this problem.

2.1. Graft Copolymers

Graft and block multiphase copolymers have received an
increasing amount of attention over the past twenty years.
They have been shown to possess desirable combinations of
physical and chemical properties that allow them to be
useful in a variety of applications (Nos77, Pei86, Nat88,
Tan90, Zha90, Pei94) including blend compatibilizers,
surfactants, impact modifiers, and surface modifiers.

All of the applications mentioned take advantage of the
individual properties of chemically dissimilar segments.
Most of these depend on the ability of these copolymers to
act at a particular interface. For example, these
copolymers can be added to a blend of two immiscible
homopolymers that have affinities for the individual
segments of the multiphase copolymer. The copolymer
migrates to the interface between the two phases (Shu90) ,
reducing the interfacial tension (Gai80, Ana89) and thus
enhancing the strength of the interface (Fay82, Bro89,
Cre89) .

2 . 1 . 1 . Synthetic Routes to Graft Copolymers

One of the most widely used methods of synthesizing
multiphase copolymers is through sequential anionic block

copolymerization (Has83) . But the number of monomers that
can be polymerized anionically is limited due to unwanted
side reactions (Has83) . The stringent polymerization
conditions required for the anionic polymerization of
certain monomers (Nos77) also limits its applicability.
These limitations severely reduce the applicability of this
route to a wide variety of combinations of block segments.
In order to expand the possible combinations of monomers
used as the two phases, graft copolymerizations have been
investigated (Mei73) .
2 . 1 . 1 . 1 .Anionically polymerized macromonomers

One of the most prevalent and reliable synthetic routes
to graft copolymers is through the use of macromonomers

(Cor84, Rem84b, Muh87, Mei90, Pei94). Macromonomers are end
functional macromolecules capable of further polymerization

(Rem84a) . They are generally synthesized through the

anionic polymerization of a monomer followed by reaction of

the living anion with an end capping agent such as

methacryloyl chloride, producing a methacrylate or vinyl

terminated macromonomer (Mas82, Sch82, Ham84, Sch84a,

Sch84b, Cam85, Cam86, Gna87, Gna88) . Once again,

application of this synthetic procedure is limited in scope

due to monomer restrictions in anionic polymerizations.

2 . 1 . 1 . 2 .Macromonomers through chain transfer
functionalization

A more versatile route to macromonomers involves the

free radical polymerization of a monomer in the presence of

a functional chain transfer agent (Ito77). Functionalized
mercaptans are commonly used. Mercaptans are compounds
containing a sulfhydryl, -SH, functional group. They are
commonly used to control molecular weight in commercial
polymerization reactors (Ros82).

The mechanism by which f unctionalization can occur is
depicted in Figure 2.1. Steps 1 and 2 are typical process
of free radical initiation. When exposed to heat, the AIBN
breaks down into free radicals and nitrogen gas is evolved.
The AIBN radicals can thus initiate the polymerization of
vinyl compounds. If there is no chain transfer agent
present, the polymerization continues until termination by
disproportionation or combination occur. In the presence of
a mercaptan, termination can occur through chain transfer.

The hydrogen from the sulfhydryl group of the mercaptan
is readily extractable. A propagating polymer chain can
thus react with the mercaptan (Step 3) , terminating
propagation and leaving a sulfur radical on the mercaptan.
If the concentration of mercaptan is high, the mercaptan
itself can react with the AIBN radical (Step 4), also giving
a sulfur radical. The resulting sulfur radical can then
initiate the free radical polymerization of a vinyl monomer
(Step 5) .

AIBN

, CH 3 -C-N=N^-CH 3 -A* 2CH 3 ^ N + N 9
CH 3 CH 3 CH 3 2

C s N monomer polymerization
ch 3 -<:. + =i -►

CH 3 R

•"^lA^VVSA^^i^VAV-WVW^-JVAVAW^- #

r=N mercaptan C=N

4 CH 3 -2- + H:S-R' CH 3 -<:H + R' S .

CH 3 CH 3

5 R'S- + — [ " R

Figure 2.1 Mechanism of f unctionalization using chain
transfer agents.

with the mercaptan (Step 6) yielding a terminated
functionalized chain and another molecule of sulfur radical
which can react with more monomer (Step 5) to form a
reaction loop. The effectiveness of f unctionalization is
dependent on the chain transfer constant of the mercaptan as
well as the relative concentrations of mercaptan, monomer,
and free radical initiator (Tsu91). The AIBN concentration
is kept extremely low relative to the chain transfer agent
to minimize the number of chains initiated by the AIBN. Any
chains initiated by AIBN will be non-functionalized (see
Steps 2 and 3) .

The advantage of this chain transfer method is that it
can be used with a wide variety of vinyl monomer systems.
Macromonomers composed of any monomer which can be
polymerized free radically should be able to be synthesized
using this method. Also, macromonomers which themselves are
random copolymers also become feasible.

The resulting functionalized macromolecule can be
utilized in two different ways. As with the anionic
macromonomers, the resulting hydroxyl or carboxylic acid
monofunctionality can be converted to a methacrylate or
vinyl functionality for further polymerization with vinyl
monomers (Alb86, Che91, Tsu91) . Chen and Jones (Che91) have
synthesized hydroxyl functionalized polyacrylates using
mercaptoethanol as the chain transfer agent. Similarly,
Albrecht and Wunderlich (Alb86) have synthesized hydroxyl
terminated PMMA with Mw of 6500- 23000 g/mole. Both studies

have indicated that high levels of f unctionalization are
realized. The hydroxyl group in each case was converted to
a vinyl functionality through reaction with isocyanatoethyl
methacrylate . Tsukahara et al. (Tsu91) have synthesized
carboxylic terminated PMMA using mercaptoacetic acid as the
chain transfer agent. The acid functionality was reacted
with glycidyl methacrylate to produce methacrylate end
capped PMMA.

These routes produce methacrylate terminated polymers
which can be subsequently polymerized with a vinyl monomer
similarly to the anionically polymerized macromonomer . In
order to synthesize graft copolymers, these macromonomers
are dissolved in a solution containing a different vinyl
monomer as well as a free radical initiator. Graft
copolymers of a wide range of monomers can be synthesized in
this fashion (Che91) . But this macromonomer chemistry has
generated graft copolymers mostly limited to vinyl-vinyl
type systems (Nai92) .

The functionalized amine, hydroxyl or carboxylic acid
terminated macromonomer can also be used without the vinyl
functionalization by direct coupling with other condensable
terminated macromonomers to form block copolymers (Imi84) .
For example, amine-terminated poly (methacrylates ) and
polystyrene have been synthesized using mercaptoethyl
ammonium chloride as the chain transfer agent (Imi84,
DeB73) . Block copolymers of these macromolecules with
carboxylic acid terminated polymers were synthesized by a

coupling reaction. However, due to the monofunctionality of
the macromonomer, coupling produced only very low molar mass
A-B type block copolymers.

Another interesting use of chain transfer agents for
graft copolymerization is one developed by Moraes et al.
(Mor96) . They modified an ethylene-vinyl acetate (EVA)
copolymer by hydrolysis and subsequent esterif ication with
mercaptoacetic acid. This produced a sulfhydryl containing
polymer backbone. In essence, this is a polymeric chain
transfer agent. Methyl methacrylate (Mor96) and styrene
(Bar96) were polymerized in the presence of the mercapto-
modified EVA to give poly (EVA-g-methyl methacrylate) and
poly (EVA-g-styrene) graft copolymers.

Again, the described macromonomers and graft copolymers
have been mostly limited to addition-addition copolymers.
There are only a few cases in which addition type
macromonomers have been graft copolymerized with
condensation type monomers (Yam81, Chu82, Chu84, Chu88a) .
In order for the macromonomers to be capable of undergoing
condensation reactions in the production of graft
copolymers, they must be dif unctional . This does not imply
that each end of the polymer is f unctionalized . Instead, one
end of the polymer contains a difunctional reactive group.

Only two research groups have reported the synthesis of
difunctional macromonomers using chain transfer agents.
Nair (Nai92) has demonstrated the free radical chain
transfer polymerization of styrene and various acrylates in

the presence of mercaptosuccinic acid. Dicarboxylic acid
terminated polymers were synthesized in a range of molecular
weights from 1 to 10 kg/mole and were shown to be highly
f unctionalized.

Yamashita et al. (Yam81, Chu82, Chu84, Chu88a) carried
out brief studies on the preparation of dicarboxylic acid as
well as dihydroxyl functional macromonomers of various
methacrylate monomers. They also employed mercaptosuccinic
acid as well as thioglycerol in the macromonomer
preparation. The structures and resulting macromonomer
structures for each of the chain transfer agents discussed
are illustrated in Figure 2.2.

The work by Yamashita is the most closely related to
our synthetic approach: the evaluation of cysteine as a
chain transfer agent in the synthesis of amino acid
terminated macromonomers. Our survey of the literature
found no mention of amino acid terminated macromonomers.
2 . 1 . 1 . 3 .Modeling chain transfer reaction

The first step in the use of chain transfer agents is
the determination of the chain transfer constant. Knowledge
of the chain transfer constant allows us to predict such
critical variables as molar mass and functionality (Nai92).
The chain transfer constant can be determined using the Mayo
model (Ros82, Nai92) . The variation of the degree of
polymerization for a free radical chain transfer
polymerization is given by

Mono functional Chain Transfer Agents

HS— CH 2 -C-OH »►

HS— CH 2 -CH 2 -OH >■

CH 3

CH 2
reactive methacrylate

\\S CH^~CH? — NHolHCl &* lAMAAAftWWWSAAA/WMMAA*^^

Difunctional Chain Transfer Agents

HO— CH 2 — CH 2 -CH 2 -OH
SH

O O

HO-C— CH 2 -CH 2 -C-OH

AAAMAAWAAMMAAAMA/SWAAAAWrtAAWWWWVW

OH
COOH

AWMW*WIWWM*W«««M«W»WW

COOH

Figure 2.2 Schematic illustration of utilized chain
transfer agents and macromonomers thereof

f K *,

DP,,

+ Cs

(2.1)

where Cs = Chain transfer constant = k tr ansfer/k p
[S]= Chain transfer agent concentration
[M]= Monomer concentration
DP n = Degree of polymerization
k t ,k p ,Rp= rate constant for termination,
propagation, and the rate of polymerization
respectively .

The degree of polymerization in the absence of chain
transfer to transfer agent, DP no , can be described by
eguation 2.2 where

DP,,

k 2 p [Mf

(2.2;

Substituting this value in equation 2.1 gives us the Mayo
equation (2.3) for prediction of the chain transfer constant
where

DP,, DP,,

1 +r [s]

[M]

2.3:

This equation is valid only when the initiator
concentration is low. By synthesizing a series of polymers
with different ratios of chain transfer agent to monomer, a

Mayo plot can be used to determine the Cs . Knowledge of the
Cs for a particular system enables one to adjust the
reactant concentrations in order to target a specific molar
mass polymer.

The Mayo model can also be used to predict the
functionality of the polymer obtained. If we multiply both
sides of equation 2.3 by DP n , we get

DP,, [S]

\ = =- + DP, l Csj-j- (2.4)

DP,,,, [ M\

where the two terms on the right represent the fraction of
unfunctionalized and f unctionalized chains. If the value of
Cs and therefore the rate of chain transfer is high,
termination occurs primarily by chain transfer and high
rates of functionalization are expected. The extent of
functionalization also increases with increasing mercaptan
content. If a lower concentration of chain transfer agent
is used or if a lower value of Cs is observed, the
probability of termination through other methods such as
disproportionation or combination increases. As other
termination mechanisms become more prevalent, the extent of
functionalization decreases.

It is important to note the limitations of the Mayo
model. This model is valid only under certain assumptions
(Ath77, Nai92) . The first of these is that chain transfer
occurs exclusively to the chain transfer agent. In

practice, however, some chain transfer to solvent and
initiator is generally observed. Also, these values, as in
the case of copolymer reactivity ratios, are valid at
instantaneous conditions. In other words, low conversions
are desired in order to limit the composition drift between
the monomer and chain transfer agent. With these
assumptions in mind, determined values of Cs and predicted
functionalities are only estimates or theoretical
predictions assuming ideal conditions.

2 . 1 .2 . Polyamide Addition-Condensation Multiphase Copolymers

A variety of polyamide containing graft and block
copolymers are described in the literature. Our specific
interest lies in the ability to graft or block copolymerize
polyamides with addition polymers such as methacrylates and
other olefinic monomers. Several approaches have been taken
in the synthesis of these structures. The benefits and
limitations of each approach are described herein.
2 . 1 .2 . 1 .Poly (amide-b-olefin) block copolymers

Anionic block copolymerizations . The first evidence in
the literature of block or graft copolymerizations of vinyl
monomers with polyamide were found in the patent literature
(Fur63, Bak65, God69) . The synthetic approach taken in
these investigation involved the sequential anionic
polymerization of a vinyl monomer and an isocyanate.
Specifically, Godfrey (God69) showed that anionically

polymerized 'living' polystyrene, polyisoprene, and
poly (methyl methacrylate) could initiate the polymerization
of butyl isocyanate. The resulting product can be thought
of as a N-butyl nylon 1-b-olefin block copolymer. The
mechanism of block copolymerization is illustrated in Figure
2.3. High molar mass block copolymers were formed with
polydispersities ranging from 1.2 to 1.6. Although this
reaction was successful, the choice of polyamide in this
synthesis is limited due to the isocyanate precursors. All
block copolymers involving isocyanates will form nylon 1
type polyamides. Also, the choice of olefinic monomer is
also limited to the previously mentioned limitations of
anionic polymerizations.

Macroinitiators for lactam polymerization . The
majority of the literature concerning polyamide block
copolymers involves the anionic copolymerization of
caprolactam (Pet79, Ste82, Bor88, Mou93, YnM94), the
precursor to nylon 6. The first step in these
copolymerizations is the synthesis of polymeric
macroinitiators from the desired vinyl monomer. The end
group must be suitable for the initiation of the anionic
polymerization of caprolactam. The most recent example
involves the block copolymerization of amine-terminated
butadiene nitrile copolymer (ATBN) with caprolactam (YnM94).
The amine group is reacted with terephthaloyl biscaprolactam
to form a polymeric activator. Upon addition to a

Bu: Li

.-t

Anionic 'living' polymer
Bu— (CH 2 -CH)x-CH2-CHfLi +

k k

Bu— (CH 2 -CH)x-CH2-CH: Li +

k k

isocyanate

()

C=N -

Amide- Vinyl
block copolymer

II
hCH 2 — CH--C-N —

r * R- y

Figure 2 . 3

Mechanism of block copolymer formation through
sequential polymerizations of vinyl monomer and
isocyanates .

caprolactam solution containing additional initiator, A-B-A
type block copolymers are formed with a ATBN center block.
The molar masses of the resulting copolymers were not
evaluated. Studies of the block copolymers concentrated on
their microstructure and mechanical properties.

Similar block copolymerizations were run using ester
terminated polystyrene and isocyanate terminated
polybutadiene (Pet79) and isocyanate terminated
polyisobutylene (Won82). Both end groups can react with
caprolactam to form a macroinitiator for the polymerization
of nylon 6. A schematic of this macroinitiator formation
and block copolymerization is illustrated in Figure 2.4.
Spectroscopic evidence of A-B block copolymer structure is
shown. However, signif icanthomopolymer formation, i.e., >
30%, was observed due to coupling reactions between two
functionalized macroinitiator molecules.

Although some success has been demonstrated using the
macroinitiator method, several limitations exist. One
limitation of this approach is that this mechanism is
restricted to the polymerization of lactam based polyamides.
More importantly, Stehlicek (Ste77) and Hergenrother (Her74)
have reported that synthetic approaches that employ the
anionic polymerization of lactams activated by
macroinitiators are handicapped by the occurrence of side
reactions that may yield insoluble, crosslinked product.

2 3

Hydroxyl terminated Hexamethylene diisocyanate Isocyanate terminated

polymer polymer

° o o o

1 _^h 2 ^H^)H C=N-<CH 2 ) 6 -N=C - ^CH 2 -CH^O Jj-NH-CCH^-N J

■1

Caprolactam
o o o

2 -^CH 2 — CH^O-C-NH-(CH 2 ) 6 -N=<!! + NH— C

(CH&

I X &

Caprolactam terminated Ny)on fi b , Qck copolymer

o P ol y mer o o

3 -(C^-CH^JI-NH-tCH^-NHJ-N-^ + NH ^I J^ _^^HVAH- ( CI4)5^-k

Figure 2.4 Reaction schematic for macroinitiator formation
and subsequent anionic block polymerization of
caprolactam.

Functional polyamide macroinitiators . The synthetic
routes for block copolymerization previously described
utilize anionic polymerization methods. One interesting
twist on the macroinitiator method is to synthesize
polyamide macroinitiators (Cra80, Cra82, Den89) . These
polyamides contain functional groups which, under
stimulation from light or heat, can dissociate into
macroradicals . Thus, these macroinitiators are capable of
initiating the free radical polymerization of vinyl
monomers .

Two distinct routes have been reported. The first
(Cra80, Cra82) involves the modification of aliphatic
polyamides such as nylon 6 and nylon 6,10. The reaction
schematic for polyamide modification and subsequent block
copolymerization is illustrated in Figure 2.5. The
secondary amines in the polyamides can be nitrosated using a
variety of nitrosating agents including nitrous acid and
dinitrogen trioxide. The resulting N-nitrosoamines can
rearrange to form a diazo linkage. Diazo compounds are well
known as free radical initiators (Ros82) . The polyamide can
then, under exposure to heat (Cra80) or light (Cra82),
decompose into macroradicals and initiate the polymerization
of olefinic monomers. Block copolymers of nylon 6 and nylon
6,10 with MMA, styrene, vinyl acetate and styrene-
acrylonitrile have been synthesized by this method.

The second route to polyamide macroinitiators (Den89)
also involves the introduction of dissociative azo linkages.

Aliphatic polyamide Nitrosated polyamide

1 **wC — NH'»««««««*«wC — NH*«» + N2O3 >■ vwwC — N(NO)"*«^*«*»««»<? — N(NO)<

Diazo ester modified polyamide

■N(NO) AVW,ftwwwv,w *'C — N(NO)*

00 00

2 J_ N ( NO )- W wvwv»«v^-N(NO) ■* ^^_N=N~*~~~~»»C-0-N=N

Polyamide macroinitiators

3 AVWC O— N =N <WWWVWMMWC— O — N =N * W ~~^ *" ■ M * WWMMWW * ,< ""«"*^«"* •■W^**»W«*

-N 2 , -C0 2

Olefinic
monomer A-BandA-B-A

=1 amide-olefin block copolymer

^wwwvsw^. • . Nylon Qjeft] Olefin Nylon Olefin

Figure 2.5 Reaction schematic of block copolymerization
initiated by nitrosated polyamide
macroinitiators .

The difference is that the azo linkages are introduced
during the polyamide synthesis. Denizligil showed that
dinitriles can react with formaldehyde in the presence of
strong acids to form a polyamide. The reaction schematic is
illustrated in Figure 2.6.

Specifically, AIBN was copolymerized with formaldehyde
in the presence of sulfuric acid. AIBN is of interest
because it has both dinitrile functionality as well as a
labile azo functionality which can decompose to free
radicals under heat. The resulting polyamide was used to
initiate the polymerization of MMA and styrene in a
DMSO/methylene chloride solvent system.

Although block copolymers were formed by both processes
outline above, no account was given as to the extent of
degradation of the polyamide. These polyamide
macroinitiators function only through their ability to
degrade. That is, block copolymerization occurs through the
chain scission of the polyamide. The extent of chain
scission, especially in the AIBN based polyamide which has a
very low molar mass between azo groups, could severely deter
application of this process.

Coupling of low molar mass reactive polymers . The
synthesis of block copolymers from the coupling reaction of
prepolymerized polyamide and polyolefin with reactive end
groups has also been investigated (Mas79, Ima84, Kim92).
Synthetically, this is the least complicated method of block
copolymer synthesis. For example, Imanishi has reacted

2 7

Ns=C

AIBN

CH 3

-N=N-

CH 3

=N +

()

HCH

H 2 S0 4

Azo functional polyamide

O CH 3 CH3O

-NH 2 — C— C— N=N
CH 3

CH 3

-NH-CH 2 4

O CH 3

CH 3 O

-NH,

-N=N-

CH 3

Polyamide macroinitiator

-NH-CH 2 4

-N 2

• MVMWvVAWAVW •

Polyamide macroinitiator

Olefinic
monomer

A-B-A
block copolymer

Olefin Amide Olefin

Figure 2 . 6 Reaction schematic of AIBN containing polyamide
and block copolymer thereof.

amine terminated polystyrene with terminally haloacetylated
polyamides to produce a polystyrene-polyamide block
copolymer. Although some block copolymer is formed using
this type of coupling reaction, this method is characterized
by the highest level of homopolymer contamination and
product heterogeneity.
2 . 1 .2 .2 . Poly (amide-g-olefin) graft copolymers

The majority of the literature on amide-olefin graft
copolymers in concerned with one of two topics- the grafting
of vinyl monomers onto polyamide substrates through high
energy processes or the formation of in situ graft
copolymers at the interface of polymer blends.

Grafting onto polyamide substrates . Various methods
have been used to graft olefinic polymer chains off the
backbone of prepolymerized polyamides. The method most
often and most recently employed is through the use of high
energy processes such as UV irradiation (Bog93, You95) ,
severe oxidation/peroxidation (Ela92, Buc96a, Buc96b) ,
gamma irradiation (Mue93, Mis96) , and plasma grafting
(You95, Lee97) .

In all of these cases, grafting is surface directed.
Polyamide fibers, films, and membranes are subjected to some
form of radiation or oxidation. High energy processes are
used in order to form radical or ions at the polyamide
surface. These radicals can initiate the polymerization of
vinyl monomers to form a grafted surface.

2 9

The majority of these studies involve the grafting of
hydrophilic monomers such as acrylic acid and acrylamide
onto polyamide surfaces. One exception is the study by
Elangovan (Ela92) which has shown the grafting of PMMA onto
wool fibers through the oxidation and subsequent redox
initiation of MMA. This was done to improve the acid and
alkali resistance of the fibers. Various other applications
for these types of graft copolymers have been targeted
including the production of pH responsive membranes (You95,
Lee97, Mis96) , antibacterial fibers (Buc96a, Buc96b) , and
new media for affinity chromatography (Mue93) .

These processes are useful for their intended
application, i.e., surface directed grafting. Well defined,
isolatable graft copolymers are not synthesized by this
method. In fact, radiation grafting is usually accompanied
by significant crosslinking at the substrate surface as well
as in the graft copolymer layer (Arn97) .

In situ graft copolymers . As mentioned previously,
graft and block copolymers can be added as a compatibilizer
to immiscible polymer blends in order to improve the overall
blend mechanical properties. To this effect, the greatest
concentration of research in the synthesis of amide-olefin
graft copolymers is in the area of melt compatibilization .
This method may also be the most industrially applicable due
to the simplicity of the process as compared to the more
elaborate and labor intensive block copolymerization
methods .

Graft copolymers can be formed in situ, during the melt
blending of polyamides and polyolefins, if the polyolefin is
functionalized with a reactive group. Specific examples
include the modification of polyolefins with maleic
anhydride (Maj92, Osh92, Mod93, WuC93, Maj94a, Maj94b,
Gon95a, Gon95b, Gon95c, Sea95), glycidyl methacrylate
(Chi96) and oxazoline (Bec96) .

Glycidyl methacrylate and oxazoline have been copolymerized
with styrene. These copolymers have been blended with
polystyrene and nylon 6 to compatibilize the blend. The
randomly dispersed oxazoline and epoxide functionalities
within the styrene copolymer can react, during melt
processing, with amine end groups from the polyamide. This
results in a polyamide-g-polystyrene at the blend interface.
Both studies (Chi96, Bec96) show reduced phase size in the
blend as well as improved mechanical properties. A schematic
of this reaction is illustrated in Figure 2.7.

The most widespread use of the in situ technique for
polyamides graft copolymers involves maleic anhydride
modified polyolefins. Immiscible polyamide blends with
polyethylene, polypropylene, and S-B-S block copolymers have
been compatibilized using maleic anhydride (Maj92, Osh92,
Gon95a, Gon95b) . In general, polyolefins such as
polyethylene can be modified by grafting of maleic anhydride
in the presence of free radical initiators (Sea95). The
resulting anhydride functionality on the polyolefin can
react with polyamide end groups at high temperatures during

3 1

Poly(styrene-co-glycidyl methacrylate) Nylon 6

ch 3 o

W*WWWW« r CH? CH-T CH? (J*MMMM*MMMMMWWM<VWWVVW + NI - ^ - nC^^ C NH - " T

c=o

CH
CH?°

extrusion
+ polystyrene

Poly(styrene-graft-nylon 6) + unreacted nylon 6 + polystyrene

-£<;h 2 — ch-^— ch 2 -c

ch 3

^WMWWWWWWWWWWWWWWWWWM

c=o

H 2

H-OH

H 2

compatibilized blend

Nylon 6

Figure 2.7 Reaction schematic of in situ graft copolymer

formation from glycidyl methacrylate copolymers

extrusion of the blend. A schematic of this reaction is
illustrated in Figure 2.8.

The synthesis and application of in situ polyamide
graft copolymers has been shown to be extremely effective at
improving the phase dispersion within a variety of polymer
blends. However, these graft copolymers have not been
isolated and studied separate from the blend. This is due
in part to the crosslinking that can occur during the graft
reaction (Sea95) due to reaction of both ends of some
polyamide chains.

Macromonomer approach to amide-olefin graft copolymers .
Previous work in the area of a macromonomer approach to
polyamide graft copolymer synthesis is of particular
interest. Free radical routes such as chain transfer
f unctionalization offer more flexibility in monomer
selection than anionic macromonomer synthesis. However, as
mentioned previously, most of the previous work using either
macromonomer method has been directed at the synthesis of
vinyl terminated polyolefins and thus addition-addition
graft copolymers.

In order to synthesize graft copolymers from polyolefin
macromonomers, the macromonomer must contain a dif unctional,
condensable end group. For the synthesis of polyamide graft
copolymers, these functional groups must be composed of acid
or amine functionalities. Only Yamashita et al. (Yam81,
Chu82, Chu84, Chu88a) have reported the graft

Maleated high density
polyethylene (HDPE)

Nylon 6

rfMMMMWMMWM fc CH-) ("'{-I ) QU,--/^|-KWAMWMMAAM*rtWAM^ftWAMW -f- NH? - ttCHo)^ C NH^T"

y extrusion
+ HDPE

2 Poly(ethylene-graft-nylon 6) + unreacted nylon 6 + HDPE

-6-CH 2 — CH->— CH 2 -CH

AMAAMMMMMMMMMMVWVVMMAV

CH— CH 2
/ \

o=c

NH OH

Nylon 6

compatibilized blend

HDPE

Nylon 6

Figure 2 . 8

Reaction schematic of in situ graft copolymer
formation from maleic anhydride modified
polyolef ins .

copolymerizations of such macromonomers, dicarboxylic acid
terminated poly (methacrylates ) , with polyamide precursors.
They employed mercaptosuccinic acid (Figure 2.2) as a
chain transfer agent in the preparation of PMMA and
poly (hydroxyethyl methacrylate) macromonomers.
Macromonomers were copolymerized with aromatic diamines and
diacids to form graft copolymers. One limitation of this
approach is the stoichiometry . The degree of polymerization
in condensation reactions is controlled primarily by
conversion and stoichiometry as defined by Caruthers'
eguation (Ros82) :

\ + r

Dp n«— r (2-5:

where r is the stoichiometric ratio of reactive functional
groups. This eguation is valid under the assumption of
complete conversion.

Determining the relative concentrations of amine and
carboxylic acid groups is complicated by the introduction of
macromonomers. The amount of dicarboxylic acid added in the
reaction by Yamashita must be reduced to account for the
acid end groups on the polymer and determining the exact
amount of dicarboxylic acid becomes complicated.
Macromonomers with built in stoichiometry, that is, with one
amine and one carboxylic acid group, would be preferred.
The precise stoichiometry of amino acid terminated

macromonomers becomes especially beneficial either for
grafting with amino acids or for maximizing of graft
copolymer molar mass.

Also, only high Tg poly (methacrylate) dif unctional
macromonomers were synthesized. There is no evidence in the
literature of these type of graft copolymers containing a
low Tg rubbery phase as investigated in this study.

An interesting twist on the macromonomer approach to
amide-olefin graft copolymers was developed by Izawa et al.
(Iza93) . They synthesized vinyl f unctionalized polyamide
macromonomers which could be free radically polymerized with
vinyl monomers. A polycondensation of aromatic amino acids
was run in the presence of methacrylic acid and p-
carboxystyrene chain terminators. Complete consumption of
the macromonomers during the free radical graft
copolymerization with MMA revealed almost complete
functionalization in the macromonomers. Although this
method resulted in vinyl functionalized chains, the molar
mass of these macromonomers is about 600g/mol. This low
molar mass can be explained by the stoichiometric imbalance
caused by addition of the end capping agent.

2.2. Polymerization Shrinkage in Dental Composites

Polymer based dental composites are replacing amalgam
as the material of choice for dental restorations. The
major drive toward the use of polymer based systems is based

on the aesthetics of the restoration. Composites compare
favorably with silver amalgam in this aspect. However, the
acceptance of composites is hampered by certain property
limitations. The evolution of developments in these
materials is described below.

2 . 2 . 1 . History of Dental Composites

The evolution of dental composites represents a logical
sequence of developments based upon current technologies
well known to the non dental community. The first acrylic
filling materials were used at the time when polymer science
was a young immature science whose growth was largely due to
World War II and the need for synthetic rubber and a non
breakable canopy for fighter planes. The early polymer
based restoratives, based on methyl methacrylate monomer,
exhibited large volumetric shrinkage, low mechanical
strength, a high propensity for staining, high wear rates,
marginal leakage and inflammatory tissue responses. None of
the first generation materials had sufficient strength or
adhesion to tooth structures to withstand the rigorous
forces of oral function.

The next generation of restorative materials were
composites. The composites were based upon the
incorporation of glass particles into the resin. The glass
particles increased the mechanical strength and abrasion
resistance and reduced total volumetric shrinkage simply by

reducing the content of resin in the restoration. In
addition to the development of the composite materials, a
new reactive methacrylate type monomer was developed by
Bowen (Bow62) . Bowen's studies with epoxide resins led to a
pivotal combination of the mechanical properties of the
epoxy resin with the fast reacting methacrylate resin in the
form of BisGMA. The first BisGMA systems introduced were
polymerized by the chemical process wherein benzoyl peroxide
is combined with a tertiary amine to form free radicals at
room temperature. Later numerous variations of the light
cured systems were introduced to the spectrum of dental
materials including both UV and visible light activated
materials. The UV light was scattered by the filler
particles in the composites and thus the depth of cure was
limited. The visible light activated restorations could
achieve a greater depth of cure and thus have become the
main system.

In addition to the change in cure mechanism, numerous
examples of modifications to the BisGMA structure and
synthesis of other reactive methacrylate monomers (Lee89,
Kaw89, Joh89, Ven93) have been reported. Most of the
modifications involve either elimination of the hydroxyl
group or modifications through esterif ications or
substitution with urethane groups. There are slight
improvements in both wet and dry properties as a result of
modifications to BisGMA, however usually the differences are

3 8

not significant and many manufacturers still rely on the
BisGMA monomer for their dental composites.

The failure of these dental composites, and poor
lifetime performance as compared to amalgam restorations, is
generally attributed to a poor interfacial properties
(Sod91) . Although the glass-resin interface has been
studied extensively, the failure is usually attributed to
the interface between the composite restoration and the
remaining tooth structure.

Two of the main sources of this poor interface are as
follows :

1) the polymerization shrinkage during composite cure
causing the restoration to pull away from the remaining
tooth structure (Bau82) .

2) poor bonding between the exposed tooth structure,
composed of hydrophilic proteinaceous dentin tubules,
and the hydrophobic dimethacrylate composite (Sod91) .

The volumetric shrinkage is due to the reduction in
molar volume that occurs as vinyl monomers move from Vander
Waals distances to covalent bond distances during
polymerization. Volumetric shrinkage leads to poor marginal
adaptation to the tooth structure which causes marginal
leakage and recurrence of caries (Bra86) . Also, excessive
stresses are generated in the restoration which create
failures in both the remaining tooth structure and or the
restoration depending upon the geometry of the restoration
(Dav91). Shrinkage in current BisGMA based composite ranges

from 1 to 6 volume % (Sul93) . Variation may be attributed
more to different glass loadings and varying levels of
conversions than to any major resin development.

2 . 2 . 2 .Reduction of Polymerization Shrinkage in Dental Resins

There are many research programs focused on different
chemical structures and processes that will reduce
polymerization shrinkage (Bra92, Bye92, Sta91, Liu90). The
main thrust in dentistry has been the spiro orthocarbonate
(SOC) based systems which are the result of early pioneering
work by Bailey (Bai72). The spiro orthocarbonate reaction
involves a cationic dual ring opening mechanism which
increases the molar volume of the polymer compared to that
of the monomer. A general reaction schematic of spiro
orthocarbonate polymerization is shown in Figure 2.9.
Although zero shrinkage resins can be produced, deficiencies
with this ring opening system include the slow cure
kinetics, the inability to reduce shrinkage under non-ideal
conditions and cost of the reactive monomer (Bra92, Bye92) .
Typically, dental restorations can be cured within a few
minutes whereas the spiro orthocarbonates are very slow
reacting .

These systems do provide some insight as to a logical
step in the evolution of dental restoratives. The ring
opening polymerization kinetics may be too slow, however,

SOC monomer

r-O O-

Cationic initiator

/— O^

r -Q(~)~ r -£- R ^ ^^ r — -

Poly(SOC)

o
propagation

Y >— R R

1 <X>

O O— '

Figure 2.9 Mechanism of ring opening polymerization of
spiro orthocarbonate monomers.

the ring opening mechanism does provide a net increase in
molar volume.

The objective of this study is to determine if the
combination of the rapid kinetics of the methacrylate resin
with the ring opening of anhydride structures can be used to
minimize the polymerization shrinkage. Specifically, cyclic
anhydride functionality in the form of maleic anhydride was
be incorporated into the dental restorative based upon
propoxylated BisGMA resin.

2. 2. 3. Use of Anhydrides in Dental Applications

Anhydrides have received some attention in dental
applications. However, the anhydride is used as part of a
bonding agent and generally the ring opened form is
generally present at the time of application.

Peutzfeldt and Asmussen (Peu91) have shown that the
addition of maleic anhydride to dentin bonding agents can
increase the mechanical properties by nearly 300% when
combined with a secondary amine containing monomer such as
urethane dimethacrylate . Their studies demonstrate the
ability of the maleic anhydride to increase mechanical
properties, however, they fail to isolate the ring opening
mechanism. By mixing maleic anhydride and other anhydrides
with the primary amine or a hydroxyl containing monomer such
as hydroxyethyl methacrylate (HEMA) , either an amide linkage
or an ester linkage is formed. Thus, the ring opening

4 2

occurs prior to the polymerization reaction and hence has no
influence on the volumetric shrinkage.

Another example of the use of the anhydride structure
in dental restorative materials is 4-META or 4-
methacryloxyethyltrimellitate. Normally this reactive
component is supplied in the dicarboxylic acid form and thus
has no influence on polymerization shrinkage. It is however
very effective in promoting adhesion to the enamel structure
as well as numerous other substrates (Nak80) . The
dicarboxylic acid structure of the 4-META monomer enhances
the wetting or spread of the resin onto the tooth structure
by lowering the surface energy of the exposed structure.

2. 2. 4. Filler Modification in the Reduction of Overall
Composite Shrinkage

Methods involving reactions of the composite
reinforcing phase have been evaluated as a possible method
of offsetting polymerization shrinkage in dental composites.
Liu et al. (Liu90) have used ammonia modified
montmorillonite (MMT) as the reinforcing phase in BisGMA
composites. At temperatures between 45 and 80°C, gaseous
ammonia is released. The gas remains trapped within the
reinforcing phase and causes dilation of the montmorillonite
particles .

Liu et al. have shown that polymerization shrinkage can
be completely offset using this method in cold cure systems.
However, this process has not been extended to directly

4 3

placed dental restorations. One possible reason is that the
MMT functions due to the heat rise within the composite
system that elevates the composite temperature. The heat
rise in experimental systems using larger volumes of resin
may be greater than the heat rise seen in actual dental
restorations .

CHAPTER 3
MATERIALS AND METHODS

3.1. Materials

3 . 1 . 1 .Macromonomer Reactants

Monomers used in the macromonomer synthesis included
butyl acrylate, methyl methacrylate, and octaf luoropentyl
methacrylate . The monomers were obtained from Aldrich
Chemical Co. All were purified by fractional vacuum
distillation (l-2mm Hg) and the middle 80% was collected.
All monomers were stored under dry nitrogen over molecular
sieves .

The f unctionalizing agent used for the chain transfer
polymerization was cysteine. Cysteine was also obtained
from Aldrich and was used as received. Extreme care was
taken to keep the cysteine stored under a dry nitrogen purge
in order to prevent oxidation to cystine, the disulfide
product of the oxidation.

The solvents in the macromonomer synthesis, HPLC grade
tetrahydrofuran (Fisher), ACS grade ethanol (Fisher) and
Ultrapure™ water were used as received. HC1 was also used
in the reaction. It was obtained from Adlrich as a ION HC1
solution and diluted as reguired. Azobisisobutyronitrile

i 5

(AIBN) initiator was obtained from Kodak and purified by
recrystallization from ethanol .

3. 1.2. Graft Copolymer Reactants

Catalysts for the condensation polymerization,
triphenyl phosphite and LiCl, were obtained from Aldrich and
use without further purification. Care was taken to keep
both of these hygroscopic chemical dry. They were stored
under dry nitrogen and only opened within a drybox.

Pyridine and N-methyl pyrrolidone were used as solvents
in the graft copolymerization of the macromonomers with
polyamide precursors. Anhydrous pyridine was purchased from
Aldrich and kept continuously under dry nitrogen. Peptide
synthesis grade N-methyl pyrrolidone (NMP) was obtained from
Fisher, purified by distillation, and stored over molecular
sieves .

The amide precursors used in this study, p-aminobenzoic
acid (ABA), adipic acid (AA) , and p-phenylenediamine (PhD),
were obtained from Aldrich. ABA was used without further
purification. AA was recrystallized from ethanol-water and
PhD was recrystallized from ether.

3. 1.3. Dental Monomers

Monomers used in the modification of dental resins
included propoxylated Bisphenol A glycidyl methacrylate
(pBisGMA) , triethyleneglycol dimethacrylate (TEGDMA) , 2-

phenylethyl methacrylate (PEMA), and maleic anhydride (MA).
All methacrylate monomers were obtained from Polysciences
Inc. Maleic anhydride in the form of briquettes were
obtained from Aldrich.

TEGDMA and pBisGMA were purified by passing acetone
solutions through Aldrich inhibitor removal columns,
followed by evaporation at reduced pressure to remove the
acetone. PEMA was fractionally vacuum distilled and maleic
anhydride was recrystallized from benzene. AIBN initiator
was purified by recrystallization from ethanol .

3.2. Methods

3 . 2 . 1 . Synthesis and Characterization of Macromonomers

Synthetic procedure . Monomer concentrations in the
polymerizations were maintained constant at 16wt.%. AIBN
concentrations also remained constant at 0.1 mole % of the
monomer concentration. Cysteine levels were varied in order
to determine its affect on the polymerization of acrylates
and methacrylates .

In a typical polymerization, cysteine was dissolved in
the prescribed amount of ION HC1 in a 200ml roundbottom
flask equipped with a magnetic stirrer. Water and THF were
then added in concentrations yielding a 50g solution of
96.5/3/0.5 ratio, by weight, THF/water/HCl . lOg of monomer
were added and the desired AIBN concentration was then

dissolved in the reaction mixture. A reflux condenser was
attached to the flask. The reaction setup was then placed in
a glycerin bath at 65°C and run for 6 hours under constant
stirring. The isolation and purification of the various
macromonomers synthesized is described in the corresponding
results section 4.1.2, 4.1.3, and 4.1.4.

Characterization of macromonomers . The molar mass
distributions of the synthesized macromonomers were
characterized by GPC using a Waters HPLC system including a
Waters 600 Fluid Delivery Systems, a Waters 717 Autosampler,
and a Waters 410 Differential Ref ractometer detector. Four
Phenomenex crosslinked polystyrene columns with pore sizes
of 10 5 , 10 4 , 500, and 100A° were used in series. The flow
rate was 0.4ml/min. Sample concentrations were
approximately 0.25% in HPLC grade THF. The injection volume
was 50ul. All molar mass values were calculated using a
polystyrene calibration curve. Anionically polymerized
polystyrene standards were obtained from Polymer
Laboratories .

Transmission FTIR spectra of macromonomers were
collected using a Nicolet 20SX spectrometer. 128 scans were
collected for each sample at a resolution of 4cm" 1 . All
liquids were run between KRS-5 crystals. Solids were run in
transmission with KBr.

NMR spectroscopy was performed using a 300MHz Gemini.
The solvent used for macromonomer characterization was

■•la

deuterated chloroform. TMS was used as an internal
standard. 64 acquisitions were collected for each sample.

Elemental analysis for determination of C, H and N
content was run on an Eager 200.

ICP was run in order to determine the f unctionalization
efficiency of the macromonomer synthesis. A Perkin Elmer
Plasma 40 Emission Spectrometer was used and a wavelength of
180.73nm was monitored. This wavelength was used to
determine the sulfur content of the polymer. A sulfur
standard was obtained from Fisher and diluted using
volumetric flasks.

Sample preparation for ICP involved making a 0.5 wt . %
solution of the macromonomer in a 96/4 mixture, by weight,
of water/triton X. Triton X was used as a surfactant in
order to stabilize the emulsion of the poly (butyl acrylate)
in water. Only the liquid, low Tg, macromonomers could be
analyzed by this method. Stable emulsion of high Tg
methacrylate copolymers could not be obtained.

Sulfur content was determined in ppm in solution.
Using the value of sulfur concentration in combination with
the molar mass values we can calculate the extent of
functionalization. The following is an example of one of
these calculations. One mole of 2 . 6kg/mol poly (butyl
acrylate), from GPC analysis, in which every chain is end
capped by one mercaptan chain transfer agent residue will
contain one mole of sulfur or 32g. Therefore, 32/2600 or
1.23 wt . % . If the prepared solution contains 0.5 wt . %

4 9

macromonomer, the solution should contain 0.005*0.0123=
62ppm of sulfur. Again, if every chain were functionalized,
we should measure a sulfur concentration of 62ppm. Instead
a concentration of 46ppm was measured. From this we
estimate that 46/62 or 75% of all chains contain one sulfur
molecule or 75% are functionalized.

DSC analysis was performed using a Seiko DSC 220
interfaced with a Seiko 6500H Rheostation. The analysis was
performed at heating rate of 10°C/min under a continuous
flow, lOOml/min, of dry nitrogen gas. On average, 10 mg
samples were analyzed in crimped aluminum pans versus an
inert sapphire reference.

3 .2 .2 .Synthesis and Characterization of Graft Copolymers

Synthetic procedure . Stoichiometric molar
concentrations of amine and carboxylic acid groups were used
with a total amount of reactants egual to 5mmol . Triphenyl
phosphite was added at a 1:1 mole ratio of TPP : carboxylic
acid groups. The amount of LiCl added was kept constant
throughout as was the type and amount of solvent used.

In a typical polymerization, 1 . 37g of 2 . 6 kg/mol p(BA)
macromonomer (0.53 mmol), 0.241g (2.24 mmol) p-
phenylenediamine, 0.326g (2.24 mmol) adipic acid, 1.55g TPP
(5mmol), and 0.09g LiCl were dissolved in 30ml of an 80/20
NMP/pyridine solution in a 100ml flask. All mass readings
and component mixing was performed in a drybox. The

reaction mixture was then heated at 100°C for 4 hours. The
resulting polymer, a tacky light brown solid, was obtained
almost quantitatively by precipitation in an excess of 50/50
water/methanol nonsolvent, filtered, washed with methanol
and dried overnight under vacuum at 40°C.

Characterization of graft copolymers . Purification of
the graft copolymers was done by Soxhlet extraction using
HPLC grade THF. This was done in order to remove any
homopolymer which may result from unfunctionalized
macromonomers . Samples ranging from 0.8-1.0g were extracted
to constant weight using a Whatman cellulose extraction
thimble .

The extractant solutions were diluted to appropriate
concentrations for GPC analysis. The level of dilution
required was dependent on the amount of material extracted.
GPC was run with simultaneous detection using the
differential ref ractometer described previously as well as a
Waters 996 Photodiode Array UV detector. This detector
allows one to get a full UV scan at each elution time. This
affords the ability to determine structural differences
between UV absorbing fractions within the solution. All
other testing conditions were identical to those for
characterization of macromonomers.

Transmission FTIR spectra of the graft copolymers were
collected using KBr with collection parameters equivalent to
those described previously.

Elemental analysis and DSC methods were identical to
those used in macromonomer characterization.

The only difference in the NMR spectroscopy from that
of the macromonomer was the solvent employed. Deuterated
sulfuric acid was used as the solvent and the solvent peak
from residual undeuterated acid was used as an internal
standard.

The synthesized graft copolymers were blended with
commercial extrusion grade Nylon 6 from BASF and with blends
of Nylon 6 with 65kg/mol poly (butyl acrylate) synthesized in
this laboratory. Initial attempts were made to blend the
polymers by dissolution in dichloroacteic acid followed by
coprecipitation in methanol. After vigorously drying the
resulting powders, films were compression molded at 230°C.
This method was abandoned when the pure nylon 6 prepared in
this manner showed severe embrittlement . Either residual
acid caused degradation or the dissolution-precipitation
step removed a stabilizer.

The materials were then mixed in the solid state. In
order to get the most homogeneous mixture of graft copolymer
with Nylon 6 and graft copolymer with Nylon 6/Poly (butyl
acrylate), the samples were mixed at cryogenic temperatures.
A SPEX 6200 Freezer Mill was used at maximum impact
freguency with the sample immersed in liguid nitrogen. All
samples were milled for 8 minutes. The resulting powders
were homogeneous in appearance. All blends were compression
molded at 230°C between Teflon® coated polyimide films and

allowed to cool in the mold. The resulting films were
approximately 0.2mm thick.

Tensile properties of the films were measured using an
Instron 1122 equipped with an 890 Newton load cell at
ambient conditions. Five samples were tested for each
material according to ASTM D638M.

3_- 2 . 3 . Synt hes i s and Cha ract er i zation of Anhydride Modified
Dental Resins

Synthesis and sample preparation . An experimental
matrix was prepared composed of varying anhydride, pBisGMA,
and TEGDMA concentrations. Dental resin samples were
prepared by dissolving the maleic anhydride in the dental
methacrylate monomer compositions. 0.4 wt . % AIBN initiator
was added to the solution. The resulting viscous solutions
were transferred to a 2mm thick mold. The mold consisted of
two glass plates lined with Teflon® coated polyimide film
and a f luoropolymer elastomeric tubing to keep the solution
in the mold. The solutions were cured under a dry nitrogen
atmosphere at 75°C for 12 hours followed by a postcure at
160°C for 2 hours.

Linear copolymers of maleic anhydride with phenylethyl
methacrylate were synthesized in bulk. The desired
anhydride amount was dissolved in PEMA in a 15ml glass test
tube and 0.4 wt . % AIBN was added. The solution was purged
with dry nitrogen, sealed, and polymerized at 75°C for 4
hours. The resulting polymer was isolated by dissolution in

5 3

chloroform and precipitation in ether. Samples were filterd
and dried overnight under vacuum at 40°C.

Characterization of anhydride copolymers . Equilibrium
water content was measured using Ultrapure water. Samples
were approximately 2mm x 5mm x 10mm. Samples were placed in
an incubator at 37°C until a constant weight was reached.
The swelling solution was then changed, adding fresh
Ultrapure water and weight again monitored to insure all
unreacted anhydride has been extracted. On average
equilibrium was reached after 1-2 weeks.

All weights were measured using a Denver Instruments A-
200DS with readings taken to the fifth decimal place.

Density measurements were taken using a Mettler 33360
Density determination apparatus in combination with the
Denver Instruments scale. The Archimedes' method was used
and measurements ere taken at 25°C using Ultrapure® water.
All samples were re-weighed after testing to insure that the
time scale of these measurements were insufficient to allow
the samples to absorb water.

UV absorption spectroscopy was run on the extractant
solutions in order to monitor the extraction of maleic
anhydride. Actually, any extracted anhydride would be
extracted as maleic acid due to hydrolysis. A calibration
curve for absorbance versus concentration was made using
standards. Standards were prepared in the expected
concentration range by hydrolyzing and dissolving maleic
anhydride in Ultrapure® water. A wavelength of 274nm was

used. Thus, by monitoring the absorbance at 274nm of the
extractant solutions, concentrations of maleic acid could be
determined. Knowing the volume of water used in the
swelling experiments, we can calculate the mass of maleic
anhydride extracted. Assuming all free anhydride was
extracted, the anhydride not extracted is assumed to be
incorporated.

GPC, FTIR, and DSC of the linear maleic anhydride
copolymers were run with parameters as described in the
characterization of the polyamide graft copolymers.

CHAPTER 4
RESULTS AND DISCUSSION

4.1. Synthesis and Characterization of Amino Acid-terminated
Poly (acrylate) Macromonomers using Chain Transfer

Chemistry

The most prevalent synthetic route to well defined
graft copolymers is through the use of macromonomers, low
molar mass polymers containing a polymerizable end group. A
review of the literature has shown that current studies,
including those within this laboratory, have concentrated on
synthesizing macromonomers which are polymerizable through a
vinyl functionality. That is, they are polymerizable only
with addition type monomers through a free radical
mechanism. Because the macromonomers themselves are
generally synthesized through either anionic or free radical
polymerization in the presence of a f unctionalizing agent,
the resulting graft copolymers which can be synthesized
through this method are generally limited to addition-
addition graft copolymers such as poly (styrene-graf t-
acrylate) , poly (acrylate-graf t-methacrylate) , etc...

It was the objective of this study to synthesize
condensation polymerizable macromonomers, specifically amino
acid terminated macromonomers, capable of reacting with
amine and carboxylic acid groups in the synthesis of

polyamide graft copolymers. An amino acid functionality is
preferred over other end groups such as diacids or diamines
due to the inherent stoichiometry that it provides . This
stoichiometry is reguired in the condensation graft reaction
to insure the highest degree of polymerization possible.

The approach taken in this study involves the free
radical polymerization of a vinyl monomer in the presence of
a functional chain transfer agent (Ito77). Mercaptans,
compounds containing a sulfur-hydrogen bond, are commonly
used in chain transfer reactions. In fact, mercaptans are
commonly used to control molecular weight in commercial
polymerization reactors (Ros82).

The mechanism by which functionalization can occur is
depicted in Figure 4.1. Steps 1 and 2 are typical processes
of free radical initiation. When exposed to heat, the AIBN
breaks down into free radicals and nitrogen gas is evolved.
The AIBN radicals can thus initiate the polymerization of
vinyl compounds. If there is no chain transfer agent
present, the polymerization continues until termination by
disproportionation or combination occur. In the presence of
a mercaptan or other chain transfer agent, termination can
occur through chain transfer. The hydrogen from the
sulfhydryl group of the mercaptan is readily extractable. A
propagating polymer chain can thus react with the mercaptan
(Step 3) , terminating propagation and leaving a sulfur
radical on the mercaptan. If the concentration of mercaptan

AIBN

, CH 3 -?-N=N-^-CH 3 -A_^ 2CH 3 ^
CH 3 CH 3 C

C "N monomer polymerization

! V* +

' 3

CH3 R

^^*wwvw^**N»^S'Si«AN»ii»»«^Ai'vvviAf\'w 1 wi«vSi'

r s]sj mercaptan C =N

CH 3 -C- + H:S-R* CH 3 -^H + R - S .

CH 3 CH 3

^MWVWV>WWAA^^^^V^M^«V .

5 R'S« + "H R'

Figure 4.1. Mechanism of functionalization using chain
transfer reactions.

is high, the mercaptan itself can react with the AIBN
radical (Step 4), also giving a sulfur radical. The
resulting sulfur radical can then initiate the free radical
polymerization of a vinyl monomer (Step 5) .

If the mercaptan contains hydroxyl or carboxylic acid
functional groups (R' ) , the initiating sulfur radical
introduces functionality to one end of the macromolecule .
The growing f unctionalized polymer radical can again react
with the mercaptan (Step 6) yielding a terminated
f unctionalized chain and another molecule of sulfur radical
which can react with more monomer (Step 5) to form a
reaction loop. The effectiveness of f unctionalization is
dependent on the chain transfer constant of the mercaptan as
well as the relative concentrations of mercaptan and free
radical initiator (Tsu91) . The AIBN concentration is kept
extremely low relative to the chain transfer agent to
minimize the number of chains initiated by the AIBN. Any
chains initiated by AIBN will be non-f unctionalized (see
Steps 2 and 3) .

Cysteine, a naturally occurring amino acid containing a
sulfhydryl functional group, was evaluated as a chain
transfer agent in the polymerization of acrylates and
methacrylates . If cysteine were to act as an effective
chain transfer agent, it would provide the desired amino
acid functionality.

Specifically, poly (butyl acrylate) and poly (methyl
methacrylate-co-octafluoropentyl methacrylate) macromonomers

were synthesized in the presence of cysteine. Poly (butyl
acrylate) was chosen because it has a very low glass
transition temperature, -54°C (Bra89), and therefore any
graft copolymers containing it may be used as rubber
modifiers. The f luoroacrylate copolymer was chosen because,
due to the low surface energy of f luoropolymers in general,
graft copolymers could be utilized as surface modifiers.

4 . 1 . 1 . Determination of an Appropriate Solvent System

A common solvent for both the monomer, either butyl
acrylate or the f luoroacrylate-MMA mixture, and the cysteine
chain transfer agent must be identified in order to insure a
homogeneous solution during polymerization. Cysteine is a
crystalline powder insoluble in common organic solvents.
Solubility tests in the approximate concentrations required
for synthesis of a 3kg/mole macromonomers were performed.
As is shown in Table 4.1, at the appropriate concentrations,
cysteine is insoluble in some common organic solvents which
are suitable for the polymerization of butyl acrylate. Also
butyl acrylate is completely immiscible with water, quickly
separating into two layers.

6 (J

TABLE 4 . 1 Solvent determination for monomer and chain
transfer agent

Toluene

THF

DMF

Ethanol

n-Butanol

H 2

Cysteine

Butyl
acrylate

Concentration of cysteine= 0.03g/5g solvent,
lg/5g solvent. i= insoluble, s=soluble.

Concentration of monomer=

Because of the strong H-bonding interactions within the
amino acid, it appeared to be necessary to add H 2 to
disrupt crystalline structure. Cysteine was then pre-
dissolved in water at high concentrations prior to the
addition of THF. This method was successful in keeping
cysteine dissolved in a THF/water mixture. One of two
things generally occurred upon the addition of butyl
acrylate. Either the concentration of water was too high to
allow the butyl acrylate to dissolve, or the concentration
was too low to prevent the precipitation of cysteine upon
the addition of the acrylate monomer. It was determined,
after much trial and error, that ethanol could be added in
low concentrations in order to stabilize the

THF/water/cysteine/butyl acrylate solution. The ethanol was
effective in preventing the butyl acrylate from forming a
second phase. The composition of the solvent system used
was a 80/10/10 ratio, by weight, of THF/EtOH/H 2 0. The
monomer concentration was 15 wt . % .

6 1

4 . 1 . 2 . Preliminary Studies of the Effectiveness of Cysteine
as a Chain Transfer Agent in the Polymerization of Butyl
Acrylate

The synthesis of poly (butyl acrylate) in the presence
of cysteine was carried out using the solvent system
described above. Figure 4.2 depicts the desired amino acid
functionality of the macromonomer . The

monomer : cysteine :AIBN molar ratio used was 1000:30:1. The
AIBN concentration must be kept low in order to minimize the
number of chains initiated by AIBN. As stated previously,
any chains initiated by the AIBN initiator and not the chain
transfer agent will be in effect 'dead' chains. That is,
they will lack the desired amino acid functionality. The
polymerization was run under nitrogen at 65°C for 7 hours.
A control polymerization was also run under the identical
conditions in the absence of the cysteine chain transfer
agent. The resulting polymers were isolated by rotary
evaporation under vacuum at 40°C. Due its low glass
transition temperature, poly (butyl acrylate) is virtually
impossible to isolate by precipitation in a non-solvent.
The reaction product of the control reaction was a clear,
extremely tacky, viscous material with a yellowish haze.
The cysteine modified product was very similar with the
exception of the presence of a white precipitate dispersed
within the poly (butyl acrylate). This precipitate could be
separated from the polymer by dissolving the poly (butyl

butyl acrylate

CH2=CH

U
i

CH 2
CH 2
CH 2

CH,

cysteine

NH 2 -CH-C-OH
CH2

0.2wt%AIBN t
65°C, 7 hrs.

THF: EtOH :H 2

O
NH 2 — CH— C-OH

CH 2

Poly(butyl acrylate)

Figure 4.2 Schematic of ideal amino acid f unctionalization
during polymerization of butyl acrylate.

6 3

acrylate) in THF. The precipitate was insoluble in THF and
could therefore be collected by filtration.

Gel permeation chromatography of p(BA) . Molar mass
averages and distributions were measured using GPC in order
to determine the effectiveness of chain transfer. Chain
transfer should significantly reduce the molar mass of the
resulting polymer. If cysteine were to have a chain
transfer constant equivalent to other commonly studied
mercaptans, it should drastically decrease the molar mass of
p(BA) when compared to a neat polymerizaion . For example,
similar concentrations of mercaptoethanol chain transfer
agent have been used in this laboratory in the synthesis of
hydroxyl terminated poly (styrene-acrylonitrile) . Using the
Mayo equation described in section 2.1.1.3, we can calculate
that mercaptoethanol (chain transfer constant=l . 1 (Zam95))
would yield an oligomer with a molar mass of @4kg/mol.

The molar mass distributions of the control and
cysteine modified P(BA) are shown in Figure 4.3. The number
average molar mass (Mn) for the control sample was 63kg/mol,
with a polydispersity index of 2.9. The product of the
cysteine modified polymerization has an Mn= 19kg/mol. The
molar mass was reduced, but not as much as would be expected
if efficient chain transfer occurred. Also, the cysteine
modified P(BA) has a much broader distribution, with a
polydispersity index of 4.8.

Analysis of GPC data . In order to properly explain
these results, the presence of the side product of the

1 -

neat Poly(butyl acrylate)
Mn=63Kg/mol Mw=186Kg/mol
PDI=2.9

PBA-cysteine pH=6.7
Mn=19Kg/mol Mw=91Kg/mol
PDI=4.8

5 4

LogMW

Figure 4.3

GPC results of preliminary butyl acrylate
polymerizations .

cysteine reaction must be explained. The side product, with
a melting point of @ 196°C, was determined to be crystalline
by DTA (Figure 4.4). The FTIR spectra of this product,
shown in Figure 4.5, shows the presence of a ester carbonyl
at 1740cm" 1 . The broad absorption from 3400 to 2400cm" 1
suggests the presence of a carboxylic acid functionality.
Figure 4.6 shows the same spectra of the side product,
focusing on the area from 1800 to 1500cm" 1 , as compared to
butyl acrylate monomer. The side product does not have the
sharp absorbance at 1640cm" 1 associated with the vinyl
functionality in the butyl acrylate monomer. There is a
broader absorption centered around 1580cm" 1 which is
representative of an amine functional group.

Upon investigation of the literature for possible
explanations of this side reaction of cysteine, a study by
Friedman et al. (Fri65) was found in which they investigated
possible blocking agents for sulfhydryl groups in proteins.
They showed that acrylates, in aqueous conditions, can react
with cysteine through the sulfhydryl group. The S:H
functional group can ionize under basic conditions into a
sulfur anion according to their reaction pathway (Figure
4.7). The sulfur anion then attacks the vinyl group of the
acrylate. The resulting carbanion is immediately capped by
a proton in the aqueous solution.

50 100 150 200 250 300 350 400

Temperature (°C)

Figure 4.4 DTA trace of side product

1.0

-Q

if)

£1
<

0.4 -!

3000 2000

Wavenumbers (cm-1)

1000

Figure 4.5 FTIR spectra of side product of cysteine
modified P (BA) .

1.0 1

-j ester carbonyl
0.8 \ \

CD
.Q

C/5
-Q

Side product of cysteine reaction

amine

1800

1700 1600

Wave numbers (cm-1)

1500

Figure 4.6 Comparison of FTIR spectra of side product with
butyl acrylate monomer.

RSH + H 2 ^ RS- + H 3 +
RS- + CH 2 =CH-COOR ■» RS— CH 2 — CH-COOR
RS—CH 2 — CH-COOR + H 3 + -» a RS-CH 2 — CH 2 -COOR + H 2 Q

Figure 4.7 Reaction pathway of cysteine with acrylates.

When butyl acrylate is used, the product formed is
S-carbobutoxyethylcysteine (Figure 4.8). According to
Friedman, this compound has a melting point of 194-195°C,
which is in good correlation to the T m of the side product,
i.e., 196°C. Elemental analysis of the side product
confirmed the correct chemical formula for S-
carbobutoxyethylcysteine (Figure 4.8).

In order to minimize or eliminate this side reaction,
we must understand why it is occurring. In an aqueous
solution, the ionization of the sulfhydryl group is an
equilibrium reaction. It is the anionic form of cysteine
which can react with butyl acrylate to form the side
product. This equilibrium reaction has a pK a associated
with it and therefore the relative concentrations of ionized
to unionized species are influenced by the pH of the
solution as governed by the Henderson-Hasselbach equation,

pH = pK a + log

[SH].

4.i;

where [S-] is the concentration of sulfur anion and [SH] is
the concentration of the unionized sulfhydryl group. The pK d

S-Carbobutoxyethylcysteine

Nhfe-CH-C-OH
6Hz
S

6H2

(Chfe) 3

CHs

Calculated wt.% for C 10 H 19 NO 4 S:
C, 48.19; H, 7.63; N, 5.62

Found:

C, 48.47; H, 7.65; N, 5.51

Figure 4.8 Structure and elemental analysis of side
product .

7 .1

of the sulfhydryl group of cysteine is 8.3 (Voe90). In
order to determine the relative concentration of ionized
cysteine, the pH of the THF/EtOH/H 2 0/butyl acrylate/cysteine
solution was then measured. The pH of the solution, 6.7, is
below the pK a . Using equation 4.1, we can then determine
that approximately 3% of the cysteine at a pH of 6.7 is
present in the ionized form and therefore 3% is capable of
reacting with butyl acrylate in the side reaction.

If the effective concentration of chain transfer agent
were only being reduced by 3%, we would expect that there
would be sufficient unionized active cysteine to
significantly affect the molar mass of the poly (butyl
acrylate) . Again, in these chain transfer polymerizations,
the molar mass is governed by the chain transfer constant as
well as the relative concentrations of monomer, chain
transfer agent, and free radical initiator. A 5% decrease in
cysteine concentration would indicate that instead of a
1000:32:1 monomer : cysteine :AIBN molar ratio, we would be
working with a 1000:31:1 molar ratio which should still
significantly decrease the molar mass of the poly (butyl
acrylate) . One would expect a 5kg/mole instead of 4kg/mole
macromonomer .

The final mixture of products in this reaction depends
upon the relative rates of cysteine consumption in the chain
transfer reaction and the side reaction. Friedman has shown
that the reaction between the sulfur anion and acrylates
occurs almost instantaneously (Fri65) . On the other hand,

the polymerization of butyl acrylate occurs over a period of
hours. The ionized cysteine is thus consumed at a higher
rate than unionized cysteine. The desire for equilibrium in
the ionization reaction (Figure 4.9) drives the reaction
further to the right which leads to further ionization of
cysteine and thus formation of more side product. Over
time, the unionized cysteine concentration decreases until
there is no cysteine present to cause chain transfer.

The unexpected GPC results shown previously in Figure
4.3 can now then be explained. At the early stages of the
polymerization of poly (butyl acrylate), unionized cysteine
is present at relatively high concentrations and low molar
mass macromonomer is being produced. As time goes on, the
cysteine concentration relative to butyl acrylate is
decreasing leading to less termination through chain
transfer and therefore larger molar masses. By the end of
the reaction, all of the cysteine has been consumed by the
side reaction and high molar mass poly (butyl acrylate) is
being formed. This would explain the extremely broad molar
mass distribution observed in the GPC analysis. More
pertinent than the actual molar mass values is that in the
absence of cysteine, initiation will come mainly from the
AIBN and termination will occur by disproportionation
(Bra89), leading to high molar mass unf unctionalized chains.

Prevention of undesired side reaction . It is clear
that in order to achieve complete chain transfer and

o o

pKa=8.3
NH 2 — CH-C— OH ^ NH 2 — CH-C— OH

CH 2
SH

butyl acrylate

Chain transfer
functionalization

H +

CH 2

T butyl acrylate

S-Carbobutoxyethyl-
cysteine

Figure 4.9 Side reaction preventing complete chain
transfer.

therefore produce highly f unctionalized macromonomers, side
reactions must be reduced or eliminated. In order to
prevent the formation of S-carbobutoxyethylcysteine, we must
prevent the ionization of cysteine. By acidifying the
polymerization solution, the initial concentration of sulfur
anion can be virtually eliminated. For example, again using
equation 4.1, if we were to acidify the polymerization
solution to a pH of 1, the concentration of ionized cysteine
would be reduced from 5% to 50 parts per billion. Although
this ionized cysteine will still form the byproduct, the
effective concentration of unionized cysteine will remain
the dominant species.

A new reaction was run under similar conditions to the
previous cysteine modified polymerization with the addition
of hydrochloric acid. The addition of HC1 facilitated the
dissolution of cysteine. In fact, a homogeneous
polymerization solution could be achieved in a solvent
system containing 96.5/3/0.5 weight ratio THF/water/HCl .
Because the concentration of water could be significantly
reduced, the addition of ethanol was not necessary to keep
the butyl acrylate monomer in solution. The pH of the
reaction mixture was reduced to -0.76 and the reaction again
was run for 7 hours at 65°C. At this pH, only lppb of
cysteine is present in the ionized form.

The resulting solution was neutralized with pyridine.
Pyridine hydrochloride immediately began to precipitate and
was filtered out. The poly (butyl acrylate) solution was

evaporated to dryness under vacuum at 40°C. The product
yield was 52% of theoretical. Again a yellowish viscous
liquid was isolated. The most significant visual difference
between this product and those of the previous reaction was
the much lower viscosity, evidence of lower molar mass
polymer. Also there was no evidence of the formation of the
S-carbobutoxyethylcysteine .

GPC was run on the isolated product. Figure 4.10 shows
the effect of acidification on the molar mass distribution
of poly (butyl acrylate) . The concentration of cysteine was
identical to that in the previous reaction. Yet, the molar
mass and polydispersity of the poly (butyl acrylate) was
significantly reduced, i.e., with an Mn of 2 . 6kg/mol and a
polydispersity index of 2.3. The addition of HC1 allowed
cysteine to participate in the chain transfer reaction. As
was stated previously, if cysteine were to behave as
effectively as other mercaptan chain transfer agents, a
molar mass of 4kg/mol would be expected.

4 . 1.3. Determination of Chain Transfer Constant of Cysteine
in the Synthesis of Amino Acid- terminated Poly (butyl
acrylate) Macromonomers

The first step in the use of chain transfer agents is
the determination of the chain transfer constant. Knowledge
of the chain transfer constant allows us to predict critical
variables such as molar mass and functionality (Nai92) . The
chain transfer constant can be determined using the Mayo

1 -

neat Poly(butyl acrylate)
Mn=63Kg/mol Mw=186Kg/mol
PDI=2.9

PBA-cysteine pH=6.7
Mn=1 9Kg/mol Mw=91 Kg/mol
PDI=4.8

PBA-cysteine: HCI pH=-0.76
Mn=2.6Kg/mol Mw=6.1 Kg/mol
PDI=2.3

6 5 4

LogMW

Figure 4.10 Effect of acidification on chain transfer
functionalization of poly (butyl acrylate)

7 7

model (Ros82). The variation of the degree of polymerization
in a free radical chain transfer polymerization is given by

'*, *,

DP,,

+ Cs

4.2)

where Cs = Chain transfer constant

[S]= Chain transfer agent concentration
[M]= Monomer concentration
DP n = Degree of polymerization
kt,kp,R p = rate constant for termination,
propagation, and the rate of polymerization
respectively.

The degree of polymerization in the absence of chain
transfer, DP no , can be described by equation 4.3 where

DP n

KM 2

4.3)

Substituting this value in equation 4.2 gives us the Mayo
equation (4.4) for prediction of the chain transfer constant
where

1 +r [S]
= + Cs-,

DP,, DP,,

[M]

(4.4)

This equation is valid only when the initiator
concentration is low. By synthesizing a series of polymers
with different ratios of chain transfer agent to monomer, a
Mayo plot (Tsu91) can be used to determine Cs to prove that
cysteine is an effective chain transfer agent.

The Mayo model can also be used to predict the
functionality of the polymer obtained. If we multiply both
sides of equation 4.4 by DP n/ we get

DP,, — [S]

where the two terms on the right represent the fraction of
unfunctionalized and f unctionalized chains.

Synthesis . Butyl acrylate : cysteine :AIBN mole ratios of
1000:64:1, 1000:32:1, 1000:16:1, 1000:0:1, were polymerized
in a THF/H 2 0/HC1 solvent system identical to that previously
described in section 4.1.2. The only difference was in the
purification. After isolation, the polymers were extracted
with Ultrapure™ water in order to remove any traces of
unreacted cysteine or AIBN.

GPC results . Molar mass distributions are shown in
Figure 4.11. The Mn systematically decreases as the
relative concentration of chain transfer agent is increased.
The Mn of the neat poly (butyl acrylate) was 65kg/mol with a
polydispersity index is 3.0. The Mn at the highest
concentration of cysteine was 1.3kg/mol. The

polydispersities of the reactions in which cysteine is
present are all significantly lower, i.e., around 2.0.

A Mayo plot of the GPC data is depicted in Figure 4.12
Referring back to equation 4.4, we can plot the relative
concentration of chain transfer agent to monomer, [S]/[M],
versus the inverse of the number average degree on
polymerization. After performing a regression on the data,
the slope of the line is the chain transfer constant, Cs .
The chain transfer constant

Cs = K / kp (4.6)

where k tr is the rate constant for chain transfer. A larger
chain transfer constant indicates increasing termination by
chain transfer and thus the effectiveness of the chain
transfer agent in participating in the polymerization. The
chain transfer constant calculated for the butyl
acrylateicysteine system was 1.49. Reported Cs values for
other mercaptans are generally between 0.8 and 1.2. As
stated previously, we have studied mercaptoethanol as a
chain transfer agent and found a Cs of 1.1. Thus, cysteine
is an extremely effective chain transfer agent in the
polymerization of poly (butyl acrylate).

Spectroscopic characterization. FTIR was run in order
to observe any differences in structure between the neat
poly (butyl acrylate) and the product of the cysteine

20 -

1000:64 BAxysteine
1000:32 BAxysteine
1000:16 BAxysteine
neat PBA

5 4

log (MW)

Figure 4.11 GPC results of polymerizations of butyl
acrylate varying cysteine concentrations

12
10

8
6
4
2

1/DP n = 1/DP n + Cs[S]/[M]

Cs=1.49

r 2 = 0.99

T = 65°
THF/H20/HCI
[AIBN] = 0.2wt%
S : Cysteine

[S]/[M]x10 :

Figure 4.12 Mayo plot for chain transfer constant

determination for cysteine :butyl acrylate
system.

modified polymerizations. Spectroscopic determination and
quantification of amino acid functional groups is
complicated by the low end group concentration relative to
P(BA) .

The only difference in the FTIR spectra of the lowest
molar mass poly(butyl acrylate) macromonomer (1.2kg/mol) and
the neat poly(butyl acrylate) (Figure 4.13). The only
significant difference between the two spectra was the low
intensity broad absorption from 3400-2400cm _1 . This was
attributed the carboxylic acid from the cysteine end group.

NMR spectroscopy was used in an attempt to better prove
the presence of the amino acid end-group. A comparison of
the 1 H-NMR spectra of neat poly (butyl acrylate) (Fig 4.14)
and the 1.2kg/mol macromonomer (Fig 4.15) reveals the
presence of two very small, broad peaks around 8 2.8 and 3.3
for the macromonomer. The chemical shift values are equal
to the predicted shifts of the methylene and methine protons
from the cysteine end-group (Sil91) . Again, the peaks are
very low in intensity due to the low end group
concentration. The 13 C-NMR spectra of the low molar mass
macromonomer is shown in Figure 4.16. All peaks can be
assigned to that of poly (butyl acrylate) with the exception
of a small peak at 5 31.8 which is assigned to the C-S-R
carbon .

ICP was used to determine the sulfur concentration in
the p(BA). As described in section 3.2.1, this value in

c
n

-Q
O

1.0

0.8

0.6

0.4

Poly(butyl acrylate)

0.2 -

Amino acid

terminated

P(BA)

0.0 >-

4000

I ■ ■ ■ | < < • | ' ' I | I I I | I c I I i I ■ I • I • I I I ■ I I I I I I I I I ■ • I I I I I I I I I |

3000 2000

Wavenumbers (cm-1)

1000

Figure 4.13 FTIR spectra of poly (butyl acrylate) and
cysteine end-capped p(BA) macromonomer .
Macromonomer is the 1.2kg/mol p(BA)
synthesized using 1000:64:1 butyl
acrylate:cysteine : AIBN mole ratio.

9 4

4ch 2 -

CH-)-
x

c=o

I
CH 2

CH 2

I
CH 2

I
CH 3

TMS

«UL

1 I ' I I I < " 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

1 PP m

Figure 4.14 X H-NMR spectra of neat poly (butyl acrylate) .

8 5

0= ? g f e

hCH-CH 2 -S-(-CH 2 -CH-)-

NH-

i — i — r

c— o

CHo d

:h 2 c

ch 2 b

TMS

Kl.

i i i | i i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — p — r

PPm

Figure 4.15 X H-NMR spectra of 1.2kg/mol cysteine modidied
poly (butyl acrylate) .

8 6

o=c

h , g f ,
CH-CH 2 -sfCH 2 -CHi-

180 160 140 120 100 80 60 40 20 ppm

Figure 4.16 13 C-NMR spectra of 1.2kg/mol cysteine modidfied
poly(butyl acrylate) .

combination with the measured molar mass values allows us to
estimate the percent f unctionalization (Table 4.2).

TABLE 4.2 Percent functionalization versus molar mass for
poly (butyl acrylate) macromonomers .

[S/M]

2
x 10

Mn
(Kg/mole)

% functionalization
(ICP- S cone. )

Predicted %
functionalization*

1.6

5.9

72 ± 2

3.2

2.6

75 ± 2

6.4

1.3

84 ± 1

Functionalization= l-DP n /Dp no derived from equation 4.5. [S/M]
relative molar concentration of cysteine to butyl acrylate.

As the concentration of cysteine was increased, the
percent functionalization of chains increases up to 84% for
the lowest molar mass sample. This trend of increasing
functionalization with increasing cysteine concentration
also held for the predicted functionalities from the Mayo
model. This would imply that chain transfer dominates the
termination process as the chain transfer agent
concentration increases. Also, the probability of
initiation by an AIBN radical versus a mercapto sulfur
radical decreases.

The differences in the predicted and experimental
functionalities may be accounted for by experimental

methodology used to measure these values. As stated
previously in section 3.2.1, these values are calculated
under the assumption that the molar masses measured by GPC
are absolute, where in fact these values are relative to
polystyrene standards. In other words, our measured molar
mass is the molar mass of the polystyrene standard whose
hydrodynamic volume is equal to our sample. Nonetheless,
because of similar solubility parameters, monomer molar
masses, and backbone chemistry for poly (styrene) and
poly (butyl acrylate) , these estimates should be reasonably
accurate. It must be noted that the values predicted by the
Mayo model are also estimates calculated under the
assumption of ideal conditions.

It is reasonable to assume that the sulfur content
measured by ICP for the macromonomer is present due to the
cysteine end-group. If this is the case, then for every
mole of sulfur we have a mole of amino acid. Therefore we
are estimating that between 72 and 84% of all chains possess
the amino acid functionality and therefore the ability
participate in a condensation reaction.

4 . 1.4 .Cysteine Chain Transfer in the Synthesis of Amino

Acid-terminated Poly (methyl methacrylate : octaf luoropentyl
methacrylate) Macromonomers

Fluoroacrylate copolymers were also polymerized in the
presence of cysteine in order to produce amino acid
terminated f luoropolymers . Fluoropolymers are known for

their low surface energy. They have been blended with more
hydrophilic polymers to reduce their moisture absorption.
If f luoropolymers could be synthesized which were capable of
grafting into a polyamide backbone, we could envision this
graft copolymer could be used as a potential surface
modifier for polyamides.

The f luoropolymer chosen in this study is
octafluoropentyl methacrylate (OFPMA) . Because of concern
over the solubility of a homopolymer of OFPMA in the graft
copolymerization reaction, copolymers of this f luoroacrylate
with methyl methacrylate (MMA) were synthesized. The
copolymer structure is depicted in Figure 4.17. The MMA to
OFPMA, monomer molar masses of 100 and 300g/mol
respectively, were mixed in a 1:1 weight ratio to yield a
mole ratio of 3 : 1 MMA: OFPMA.

Monomer: cysteine :AIBN mole ratios of 1000:64:1,
1000:32:1, 1000:0:1, were polymerized in a THF/H20/HC1
solvent system. Polymerization conditions and solvent
composition were identical to those in section 4.1.2. The
only difference came in the isolation of the product. The
copolymers, which were isolated by precipitation in a 5X
excess of methanol, were filtered and dried overnight under
vacuum at 40°C. A white precipitate was isolated and then
ground into a powder.

GPC results . Molar mass distributions of the three
copolymers are shown in Figure 4.18 and the molar mass
averages are tabulated in Table 4.3. Mn values decrease

Poly(methyl methacrylate-co-octafluoropentyl methacrylate)

CH 3 CH 3

-eCH 2 -CXCH 2 — C^-

0.75 0.25

c=o c=o

I I

1 I

CH 3 H— C-H

I
H— C— F

I
F— C— F

Figure 4.17 Chemical structure of f luoroacrylate copolymer.

1000:64:1 monomer:cysteine:AIBN
1000:32:1 monomer:cysteine:AIBN
1000:0:1 monomer:cysteine:AIBN

20 -

6 5

log (MW)

Figure 4.18 GPC result of chain transfer polymerizations of
f luoroacrylate copolymers.

TABLE 4.3 Molar mass values of f luoroacrylate copolymers
from GPC.

[M] : [S] : [AIBN]

Mn
(kg/mol)

Mw
(kg/mol)

PDI

1000: 64:1

19.9

34.3

2.2

1000:32:1

27.1

51.3

2.1

1000:0:1

72.1

204. 6

2.8

M= monomer, S= cysteine

from 72kg/mol in the neat polymerization to 27 and
19.9kg/mole in the presence of varying amounts of cysteine.
The molar mass distributions also are narrower, with
polydispersities going from 2.8 in the neat sample to 2.1 in
the cysteine modified polymerizations. Although the
presence of the cysteine chain transfer agent causes a
decrease in molar mass, it does so to a much lower extent
than in the previous butyl acrylate polymerizations. Figure
4.19 shows a comparison of the effect of cysteine on the
polymerizations of butyl acrylate and MMA-OFPMA. Both
distributions shown are for the 1000:32:1 mole ratios of
monomer : cysteine : AIBN .

It is evident that rate of chain transfer in the
f luoroacrylate polymerization is substantially lower than
that in the butyl acrylate reaction. In order to quantify
this decrease in chain transfer, the chain transfer constant
for the cysteine: f luoroacrylate system must be determined.

In order to generate a Mayo plot for this reaction, the
degree of polymerization for the poly (MMA-OFPMA) must be
calculated from the GPC data. But the actual composition of
the copolymer, in terms of mole percent MMA and OFPMA, is
not known and thus the average monomer molar mass and thus
the degree of polymerization cannot be calculated. Due to
possible differences in free radical reactivity ratios, the
composition of the polymer is not expected to be equivalent
the monomer feed composition.

p(BA)

1000:32 monomer:cysteine
Mn=2.6kg/mol
p(mma-co-ofpma)
40 000: 32: monomer: cysteine
Mn=27.1kg/mol

log (MW)

Figure 4.19

Comparison of effectiveness of chain transfer
for p(BA) and f luoroacrylate copolymer.

NMR spectroscopy was used to determine the comonomer
distribution in the f luoroacrylate copolymer. The spectra
for the copolymer is shown in Figure 4.20. The mole percent
fluoroacrylate was calculated from the integral ratio of -
OCH 2 - at ~ 8 4.4 4 from the OFPMA to the -OCH 3 protons at ~ 8
3.60 for MMA. Although the feed mole ratio was 3:1
MMA: OFPMA the copolymer composition was closer to 2 : 1 with
68 mole % MMA. NMR was run on all three copolymers and no
difference in copolymer composition was observed.

Given the composition of the MMA-OFPMA copolymers, we
can calculate an average monomer molar mass and thus the
degree of polymerization of our copolymer series. This
allows us to generate a Mayo plot, shown in Figure 4.21, for
the copolymerization of MMA-OFPMA in the presence of
cysteine. The chain transfer constant for this reaction has
been calculated to be 0.26, significantly lower than the
1.49 chain transfer constant for the butyl acrylate
polymerization .

The cause of the differences between the two may be due
to the lower reactivity of methacrylates versus acrylates
(Nai94) . Nair has reported that degradative chain transfer
in less reactive monomers. That is, the rate of
polymerization decreases with increasing concentration of
mercaptans. Nair attributes this to the mutual destruction
of sulfur radicals, by combination to form disulfides, and
to slow reinitiation by the sulfur radical.

c 7

CH 3

d' d

4ch 2 -c^-4ch 2 -c-4

x y

C=0 C=0

CH 2 D

a HCF

I
CF 2

CF 2
CF 3

~| 1 1 1 1 1 1 1 1 1 1 1 1 1 r- - | " ~i 1 1 r 1 —

7 6 5 4 3

TMS

C,C

-| 1 1 1 1 1 . . . 1 r

2 1

Figure 4.20 ^-NMR spectra of MMA-OFPMA macro-monomer

1.0

0.9-
0.8-
0.7-
0.6-
0.5-
0.4^
0.3-
0.2

1/DP n =1/DP n +Cs[S]/[M]
Cs = 0.26
r 2 = 0.98

[S]/[M]x10

Figure 4.21 Mayo plot for chain transfer constant

determination for cysteine :MMA-co-OFPMA system.

Conclusive evidence of the presence of the amino acid
functionality on these copolymers was not observed in the
NMR spectra of the lowest Mn copolymer (Figure 4.20) . Also,
the FTIR spectra of the neat MMA-OFPMA copolymer and that
polymerized at the highest cysteine concentration, shown in
Figure 4.22, display no observable differences.

It must be noted that because the molar mass values of
the f luoroacrylate copolymers are substantially higher than
the poly (butyl acrylate) macromonomers, the relative
concentration of end groups is substantially lower and
therefore more difficult to detect. Also, ICP could not be
used to determine sulfur concentration because stable
agueous emulsions could not be made. This could be due to
both the hydrophobicity of the copolymer as well as the
higher glass transition temperature compared to the p(BA).
The only values of functionality available are those
predicted by the Mayo model, 62 and 73% for the two MMA-co-
OFPMA macromonomers .

These values significantly lower than the predicted
functionality of the p(BA) macromonomers from Table 4.2.
Conclusive proof of functionality will be determined by the
copolymers ability to react in a condensation grafting
reaction .

1.0

0.8

<u 0.6
o

-Q
i_
O
C/)

< 0.4

Neat Poly (MMA-OFPMA)

1000:0:1
monomer : cysteine : AIBN

Poly (MMA-OFPMA)
0.2 1 1000:64:1

\ monomer : cysteine : AIBN

0.0

ft I • I I I • I • I I I ■ I I I I I I I ■ 1 ■ I I 1 I I t 1 I I I I I I I I I I ■ I I I I I

4000 3000 2000

Wavenumbers (cm-1)

Figure 4.22 FTIR spectra of MMA-OFPMA copolymers

4.2. Poly (amide-g-acrylate) Graft Copolymers from Amino
Acid-terminated Macromonomers

Most of the macromonomers described in the literature
contain unsaturated end-groups allowing for graft
copolymerizations . Therefore, the majority of the studies
involving the graft copolymerization of macromonomers have
concentrated on the free radical graft polymerization of
vinyl monomers. This study is focused on the condensation
graft copolymerization of the previously synthesized
poly (acrylate) and poly (methacrylate) macromonomers with
polyamide precursors.

4 . 2 . 1 . Synthesis of Poly (amide-g-butyl acrylate)

A general reaction schematic is illustrated in Figure
4.23. The amino acid functionality of the poly (butyl
acrylate) reacts with the specified amino acid in a
condensation reaction producing an amide-acrylate graft
copolymer and water.

The polyamide graft copolymer synthesis was carried out
in a solvent system originally developed by Higashi (Hig80a,
Hig80b, Hig80c) for the solution polymerization of high
molar mass polyamides . This polymerization of diacids and
diamines was run in an NMP-pyridine cosolvent mixture, in
the presence of triphenyl phosphite (TPP) and LiCl . The
pyridine aids in the dissolution of the amino acid

100

NH,-

CH 2

Polyamide

-COOH + NH 2 -R— C— OH

H,0

P(BA)

Poly (butyl aery late)

Figure 4.23 Amide-acrylate graft copolymer structure

101

reactants. Triphenyl phosphite catalyzes the polymerization
by reacting with the acids and amines to remove water from
the condensation. LiCl facilitates the reaction of the
triphenyl phosphite.

The condensation mechanism, as proposed by Higashi, is
illustrated in Figure 4.24. Triphenyl phosphite reacts with
LiCl to form a triphenyl phosphonium salt (Step 1) . The
phosphonium salt then reacts with a carboxylic acid to form
a diphenyl phosphonium cation and a phenolic anion (Step 2) .
An amine can then attack the carboxylic acid, producing an
amide linkage, diphenyl phosphite, and phenol (Step 3) .

Synthesis . The synthesized graft copolymer
compositions are shown in Figure 4.25. Two different
polyamide compositions, a wholly aromatic poly (aminobenzoic
acid) (PABA) and an aromatic-aliphatic poly (phenylene
diamine-co-adipic acid) (PhDAA) , were homopolymerized and
graft copolymerized in varying ratios with a 2 . 6kg/mol amino
acid-terminated poly (butyl acrylate) macromonomer . A
control reaction was run in which a 3.2kg/mol
unfunctionalized poly (butyl acrylate) was substituted for
the low molar mass poly (butyl acrylate) macromonomer. This
unfunctionalized p(BA) was synthesized using a butyl
mercaptan chain transfer agent, which should result in an
unreactive butyl end group. PhDAA was also polymerized in
the presence of a 19.9kg/mol poly (MMA-co-OFPMA)
macromonomer .

102

i CHH3 + — O-ghO

^' Li

\\ /f^ c f[^ ^ + RCO ° H

\ //

4-H <b-/"\

O ^^

R Li + Cl-

\ //

O
R

+ R'NH2

RC-NH-R' + HO— <f \ +

Figure 4.24 Triphenyl phosphite driven amide formation,

103

Polyamide
precursors

Macromonomer
(type) (weight %) Sample ID

9
Ho-c-f Vnh 2

NH-

NH 2 HO-C-(CH 2 )4-C-OH

PABA

2.6 kg/mole

66PABA-g-33BA

P(BA)
2.6 kg/mole

33PABA-g-66BA

OH —

PhDAA

p(BA)

2.6 kg/mole

33PhDAA-g-66BA

P(BA)
2.6 kg/mole

10PhDAA-g-90BA

p(MMA-co-OFPMA)

19.9 kg/mole

33PhDAA-g-66FA

p(BA)- nonfunctional ized
3.2 kg/mole

33PhDAA-g-66UBA

Figure 4.25 Synthesized graft copolymer compositions.

104

In a typical reaction, 1.37g of 2 . 6 kg/mol p(BA)
macromonomer (0.53 mmol), 0.241g (2.24 mmol) p-
phenylenediamine, 0.326g (2.24 mmol) adipic acid, 1.55g TPP
(5mmol) , and 0.09g LiCl were dissolved in 30ml of an 80/20
NMP/pyridine solution and heated at 100°C for 4 hours. The
resulting polymer, a tacky light brown solid, was obtained
almost quantitatively by precipitation in an excess of 50/50
water/methanol nonsolvent, filtered, washed with methanol
and dried overnight under vacuum at 40°C. The molar
concentration of amide precursors and catalysts were kept
constant for all polymerizations.

The PABA containing polyamides were insoluble in all
common organic solvents. Those tried included THF, DMF,
dichloroacetic acid and trif luoroacetic acid. The only
solvent found for both the PABA homopolymer and copolymer
was concentrated sulfuric acid. This is not surprising
given the similarity of the aromatic structure in PABA to
Kevlar® polyamides. PhDAA homopolymer and copolymers were
soluble in both dichloroacteic acid and concentrated
sulfuric acid.

4 .2 .2 .Characterization of Graft Copolymers

Soxhlet extraction of graft copolymers . All polyamide
homopolymers and graft copolymers were Soxhlet extracted
with HPLC grade THF, an excellent solvent for any unreacted
acrylate or methacrylate macromonomer. Grafted polyamides

105

displayed significant swelling in the THF solution. Weight
loss during extraction is shown in Table 4.4.

TABLE 4.4 Percent weight loss from Soxhlet extraction for
polyamide graft copolymers.

Polyamide Graft Copolymer

% Weight loss

PABA

0.4

66PABA-g-33BA

not measured

33PABA-g-66BA

PhDAA

0.6

33PhDAA-g-66BA

10PhDAA-g-90BA

33PhDAA-g-66FA

33PhDAA-g-66UBA

The polyamide homopolymers display negligible weight
loss during extraction whereas the graft copolymers lost
between 44 and 80 weight % during extraction. 33PhDAA-g-66FA
shows a much higher weight loss than the corresponding p(BA)
graft copolymers. This could be attributed to the
unfunctionalized chains due to the lower chain transfer
constant and thus lower efficiency of functionalization in
the polymerization of the f luoroacrylate macromonomer . In
the control polymerization, the weight loss almost
guantitatively matches the amount of unfunctionalized p(BA)
added to the reaction. This would indicate that the
unfunctionalized p(BA) is not capable of reacting with the

106

amide precursors in a graft copolymerization . Therefore,
any incorporation of the p(BA) macromonomers can be
attributed to amino acid functionality.

Due to the significant weight loss during extraction of
the macromonomer-graf ted polyamides, it would initially
appear that a large fraction of macromonomers are
unfunctionalized and therefore cannot participate in the
condensation polymerization. GPC was run on the extracted
solutions and FTIR spectra collected of the pre and post-
extracted graft polyamides in to order better explain the
Soxhlet results.

GPC results . Figure 4.26 shows comparison of the molar
mass distributions (from the refractive index detector) of
the extracted fraction of the 33PABA-g-66PBA graft copolymer
and the original p(BA) macromonomer . The extracted fraction
is not purely unreacted macromonomer. The distribution is
bimodal, with a high molar mass component at approximately
lOOkg/mol and a low molar mass fraction near the original
macromonomer molar mass.

The GPC data was also collected using a photodiode
array detector measuring UV absorption. The UV spectra from
200-400nm for the high and low molar mass components in the
extracted solution can be seen in Figure 4.27. The spectra
can be compared with that of the original macromonomer.
P(BA) shows a single absorbance centered around 220nm
characteristic of the n -» n* transitions of carbonyl
containing compounds (Sil91) .

107

70
60
50

p(BA) macromonomer
33PABA-g-66BA

5 4

logMW

Figure 4.26 Molar mass distributions of dissolved polymer
in THF extractant solution of 33PABA-g-66BA
(refractive index detector) .

108

13
<

33PABA-g-66BA
(high mw traction)

33PABA-g-66BA
(low mw fraction)

Poly(butyl acrylate)
Macromonomer

200

300
nm

400

Figure 4.27 UV spectra of dissolved polymer in THF

extractant solution of 33PABA-g-66BA from GPC-
UV.

109

Both the high and low molar mass components in the
extractant solution contain two absorbances, one around
220nm and the other centered at 315nm. This second
absorption can be assigned to the n — > n* transition typical
of conjugated aromatic compounds.

This transition is indicative of the presence of
aromatic polyamide segments. In other words, neither of the
components in the extractant solution can be attributed to
purely unreacted poly (butyl acrylate) macromonomer . The
high molar mass component can therefore be interpreted as
amide-acrylate graft copolymer. This may be attributed
either to limited solubility of our unextracted product in
THF or to a higher local distribution of poly (butyl
acrylate) in this fraction which determines its solubility.

The presence of aromatic UV absorption in the low molar
mass component, with an average molar mass similar to that
of the original macromonomer, is more difficult to
interpret. It is believed that this fraction is a mixture
of unreacted p(BA) macromonomer, due to unf unctionalized
chains, as well as very low molar mass graft copolymer in
which the amino acid terminated macromonomer has reacted
with only a few monomer units of p-aminobenzoic acid.

Similar results were obtained for the aromatic-
aliphatic PhDAA graft copolymers. Figure 4.28 shows
comparison of the molar mass distribution of the extracted
fraction of the 33PhDAA-g-66PBA graft copolymer and the
original p(BA) macromonomer. The extracted fraction again

110

100

p(BA) macromonomer
33PhDAA-g-66BA

r\ _< t i r - * r r - * l"/''Tl'"jlf''

5 4

logMW

Figure 4.28 Molar mass distributions of dissolved polymer
in THF extractant solution of 33PhDAA-g-66BA
(refractive index detector) .

Ill

is bimodal. The high molar mass component has a larger
molar mass, around 160kg/mol, than that in the PABA graft
copolymer. This peak is also more intense than in the
previous case. This may be due to the structure of the
polyamide. Copolymers with PhDAA, an aliphatic-aromatic
amide, should be more soluble than the wholly aromatic PABA.

The UV spectra of the 160 and 6kg/mol components in the
extracted solution can be seen in Figure 4.29. Each
component shows two distinct regions of absorbance, with the
longer wavelength absorbance centered around 270nm
attributed to the n -> n* transition from the phenylene
diamine units of the copolymer. The lower wavelength of the
7i -> 7i* transition in comparison to the PABA copolymers (270
vs. 315nm) is due to the difference in electronic structure
caused by the variations in aromatic substitution.

The higher molar mass component of the extraction
displays a more intense aromatic absorption than the 6kg/mol
component, indicative of a higher concentration of polyamide
in the graft copolymer. Similarly to the PABA copolymer, it
is believed that the 6kg/mol fraction is a mixture of
unreacted p(BA) macromonomer, due to unf unctionalized
chains, as well as very low molar mass graft copolymer in
which the amino acid terminated macromonomer has reacted
with only a few monomer units of phenylenediamine and adipic
acid.

The results from GPC analysis (Figure 4.30) and UV
spectroscopy (Figure 4.31) of the 10PhDAA-g-90BA

112

200

33PhDAA-g-66PBA
(high mw fraction)

33PhDAA- g- 66PBA
(low mw fraction)

Poly(butyl acrylate)
Macromonomer

400

Figure 4.29 UV spectra of dissolved polymer in THF

extractant solution of 33PhDAA-g-66BA from
GPC-UV.

113

p(BA) macromonomer
10PhDAA-g-90BA

20 -

o -

160K

' " v -„,,/

log(MW)

Figure 4.30 Molar mass distributions of dissolved polymer
in THF extractant solution of 10PhDAA-g-90BA
(refractive index detector) .

114

200

AA-g-90PBA
mw fraction)

10PhDAA-g-90PBA
(low mw fraction)

Poly(butyl acrylate)
Macromonomer

400

Figure 4.31 UV spectra of dissolved polymer in THF

extractant solution of 10PhDAA-g-90BA from
GPC-UV.

115

are similar to the other graft copolymers. Only notable
differences between this sample and the 33PhDAA-g-66BA will
be discussed. This sample contains a higher relative
concentration of high molar mass component than the 33PhDAA-
g-66BA. With a higher poly (butyl acrylate) feed
composition, the resulting copolymer should be higher in
p(BA) thus making it more soluble in THF. This would help
to explain the increase in intensity of the high molar mass
peak as well as the large % weight loss during extraction,
80%, as compared to lower acrylate containing compositions.
The UV spectra follows similar trends with respect to the
previously described graft copolymers.

The molar mass distributions and UV spectra of the
33PhDAA-g-66FA extractables are shown in Figures 4.32 and
4.33 respectively. The extracted fraction is again bimodal
with a high molar mass component around 270kg/mol. The
relative concentration of high to low molar mass components
was the highest of all graft copolymerizations . The
starting macromonomer molar mass for the f luoroacrylate
copolymer, 19.9kg/mol vs. 2 . 6kg/mol for the poly (butyl
acrylate) graft reactions, may account for the difference in
molar mass and intensity of the high molar mass fraction.

The UV results are similar to those of the previously
described graft copolymers. The most significant difference
is the reduced intensity of the 71 -> n* transition for the
low molar mass fraction. This is due to the high molar mass
of the macromonomer. This higher molar mass macromonomer,

116

p(FA) macromonomer
33PhDAA-g-66FA

»'.

i ■ » r . «, >«> .» j »*C* * * * a • *

logMW

Figure 4.32 Molar mass distributions of dissolved polymer
in THF extractant solution of 33PhDAA-g-66FA
(refractive index detector) .

117

33PhdDAA-g-66FA
(high mw fraction)

33PhdDAA-g-66FA
(low mw fraction)

FA macromonomer
(19.9kg/mol)

200

300

400

Figure 4.33 UV spectra of dissolved polymer in THF

extractant solution of 33PhDAA-g-66FA from
GPC-UV.

118

caused by less effective chain transfer in the f luoropolymer
polymerization, should contain a larger concentration of
unfunctionalized chains.

As was mentioned previously, a control polymerization,
in which PhDAA is polymerized in the presence of an
unfunctionalized low molar mass poly (butyl acrylate), was
run under identical conditions to other grafting reactions
(Sample 33PhDAA-g-66UBA) . The molar mass distributions
(Figures 4.34) of the extractant from this reaction and the
unfunctionalized p(BA) are nearly identical. There is also
no difference in the UV spectra (Figure 4.35) of the two
polymers. As was stated previously, the percent weight loss
during extraction was almost identical to the initial feed
concentration of p(BA). Hence, the extracted fraction was
exclusively unreacted poly (butyl acrylate). In other words,
grafting occurs only when the poly (butyl acrylate) is amino
acid- terminated.

FTIR results . FTIR spectra was collected in order to
verify the presence of both the polyamide and acrylate
segments, and to monitor the change in structure before and
after extraction. Sample designation will contain an 'x'
signifying that the polymer has been purified by extraction.
For instance, 33PABA-g-66BAx designates the graft copolymer
that has been purified by extraction and 33PABA-g-66BA
identifies the neat polymer recovered directly from the
graft copolymerization by precipitation.

119

Functionalized p(BA)
PhDAA-g-66UBA

\
\

\ \

\ ^.
\ \

I k
\ V

\ \

\ s

V\ '*»

\v *'
\ V ' •

\ *

\ V

\ %

uni
331

80 -

o -

rj
A

//
//
//

1 1 1

■'»

1 L

log(MW)

Figure 4.34 Molar mass distributions of dissolved polymer
in THF extractant solution of 33PhDAA-g-66UBA
(refractive index detector) .

120

33PhDAA-g-66UBA

unfunctionalized p(BA)

200

300
nm

400

Figure 4 . 35

UV spectra of dissolved polymer in THF
extractant solution of 33PhDAA-g-66UBA from
GPC-UV.

121

In Figure 4.36, the spectrum of 66PABA-g-33BAx is
compared to that of the macromonomer and the PABA
homopolyamide . Absorbances typical of aromatic polyamides,
the N-H stretch at 3300cm" 1 , the C=0 amide stretch at
1658cm" 1 , and an overlap of the N-H bend and C=C stretch
around 1513cm" 1 , are observed for the poly (p-aminobenzoic
acid) (PABA) .

The p(BA) macromonomer displays distinct aliphatic C-H
stretch absorptions centered at 2965cm" 1 as well as the
characteristic C=0 stretch for the ester carbonyl at
1732cm" 1 . The purified graft copolymer 66PABA-g-33BAx
display absorptions characteristic of both the poly (butyl
acrylate) and the aromatic polyamide segments in the graft
copolymer. Actual fractions of each phase will not be
calculated using FTIR. Nitrogen content from elemental
analysis was used to determine the actual polyamide content
in the graft copolymers and these results will be discussed
later. It must be noted that the spectra of the unpurified
neat 66PABA-g-33BA is not shown because the entire sample
was purified by extraction. However, the spectra of both
the purified and unpurified graft copolymers will be
displayed for all of the other compositions.

Figure 4.37 shows the FTIR spectra of 33PABA-g-66BA
before and after extraction. As expected the poly (butyl
acrylate) absorptions are more intense than for the previous
sample containing only 33% poly (butyl acrylate) in the feed.
Also, there is a visible decrease in the ester carbonyl

122

: poly(p-aminobenzoic acid)

0.8 -

1513

a) 0.6
o

03
JQ

i—

o
tf)

< 0.4 4

\ p(BA) macromonomer

2965

] 66PABA-g-33BAx

0.2 \ 3300

0.0 :-
4000

1732

1513

3000 2000

Wavenumbers (cm-1)

Figure 4.36 Transmission FTIR spectra of 66PABA-g-33BAx
graft copolymer.

123

1.0

0.8^

PABA

0.6

• 33PABA-g-66BA

0.4

33PABA-g-66BAx

0.2

0.0

1658

1513

1732

7 i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | . i i ( i i i | i i i 1 1 1 1 | 1 1 i | it i-j-t-t t-|

4000

3000 2000

Vtevenumbers (cm-1)

1000

Figure 4.37 Transmission FTIR spectra of 33PABA-g-66BAx
graft copolymer.

124

intensity after extraction with THF. This would indicate
that the majority of polymer extracted is either unreacted
poly (butyl acrylate) or high volume fraction poly (butyl
acrylate) -containing graft copolymer.

The spectra of PhDAA is very similar to that of PABA,
with characteristic amide absorbances at the expected
wavelengths (Fig 4.38). The major difference between the
PABA and PhDAA is the presence of an aliphatic C-H stretch
region around 2945cm" 1 from the methylene group in the
adipic acid segments. The spectra of 33PhDAA-g-66PBA and
33PhDAA-g-66PBAx both indicate the presence of p(BA) by the
1732cm" 1 ester carbonyl stretch.

Referring back to Table 4.4, this sample showed a 42%
weight loss during extraction. Again, there is a visible
decrease in the ester carbonyl intensity after extraction
with THF indicating that the majority of polymer extracted
was either unreacted poly (butyl acrylate) or high volume
fraction poly (butyl acrylate) -containing graft copolymer.

The FTIR analysis of the highest poly (butyl acrylate) -
containing graft copolymer, 10PhDAA-g-90BA, (Figure 4.39)
contains the spectra of the neat graft copolymer, the
extracted graft copolymer and the THF soluble extrated
fraction. The 10PhDAA-g-90BA displays the expected spectra,
with a less intense amide carbonyl absorption (1658cm" 1 ) due
to the decrease of amide precursors in the feed. The
purified graft copolymer shows a decrease in poly (butyl

125

10 1 PhDAA

0.8

0.6

i 33PhDAA-g-66BA

</>

33PhDAA-g-66BAx

0.4

0.2-

0.0

i • < • i

4000

1732

| I I i | i i I | i I ■ | i I I | i . I | ■ I I | . i ■ | ■ . ■ | I ■ ■ | ■ i i | ■ • • | t-ri f I n |

3000 2000

Wavenumbers (cm-1)

1000

Figure 4.38 Transmission FTIR spectra of 33PhDAA-g-66BAx
graft copolymer.

126

1.0

0.8

c
n

0.6

0.0

i lOPhDAA-g
0.4 i -90BAx

0.2-

4000

i i i i i i i i i i i
3000 2000

Wavenumbers (cm-1)

1000

Figure 4.39 Transmission FTIR spectra of 10PhDAA-g-90BAx
graft copolymer.

127

acrylate) absorption, but still has the highest content of
p(BA) of all of the graft copolymers.

This can be seen in Figure 4.40, where the spectra of
all of the purified poly (butyl acrylate) graft copolymers
are displayed. As the butyl acrylate feed concentration is
increased, there is an increase in the intensity of the
ester carbonyl absorption (1732cm -1 ) relative to the amide
carbonyl and N-H stretch.

The last spectrum in Figure 4.39 is that of the THF
soluble fraction of 10PhDAA-g-90BAx . Going back to Table
4.4, 80 wt.% of 10PhDAA-g-90BA was extractable with THF.
This THF solution was evaporated onto a NaCl crystal for
FTIR analysis. The presence of the 1658cm" 1 amide, albeit
small, correlates with the GPC-UV results that revealed the
presence of polyamide graft copolymer within the THF
extractable fraction.

The FTIR spectra of the FA macromonomer , as illustrated
in Figure 4.41, shows intense absorptions at 1732cm" 1 for
the ester C=0 stretch and at 1170cm" 1 , which is an overlap
of the C-0 stretch of the ester and the C-F stretch
characteristic of f luoroalkanes . The spectra of the graft
copolymer, 33PhDAA-g-66FA, confirms the incorporation of the
FA macromonomers during the graft copolymerization . But the
fluoroacrylate carbonyl absorption decreases in intensity by
more than 50% after THF extraction.

This decrease was much less significant in the spectra
of the 33PABA-g-66BA and 33PhDAA-g-66BA graft copolymers.

128

Wavenumbers (cm-1)

Figure 4.40 Transmission FTIR spectra of poly (butyl
acrylate) grafted polyamides .

129

1.0"

PhDAA

1658,

§ 0.4

i i i i i i i i
3000 2000

Wavenumbers (cm-1)

Figure 4.41 Transmission FTIR spectra of 10PhDAA-g-90BAx
graft copolymer.

130

This confirmed the result from Table 4.4 in which the
FA graft copolymer shows approximately 15% higher weight
loss during extraction than the equivalent BA graft
copolymers. This behavior can be explained by the higher Cs
in the butyl acrylate macromonomer synthesis. A higher Cs
leads to lower molar mass and more highly functionalized
macromonomers which are more readily incorporated in the
condensation grafting reaction.

NMR and elemental analysis results . The compositions
of the purified graft copolymers were determined by
elemental analysis (EA) using from the nitrogen content.
Composition was also determined from the """H-NMR spectra,
shown in Figures 4.42 through 4.47, of the graft copolymers
using the integral ratios of aromatic protons from the
polyamide to methyl protons from the poly (acrylate) .

The measured graft copolymer chemical compositions are
tabulated in Table 4.5. The values calculated using NMR are
in good agreement with those from EA. Poly (butyl acrylate)
content varies from 19 to 54 wt . % for the graft copolymer
series. As expected, the poly (butyl acrylate) content in the
graft copolymer increases with increasing feed
concentration .

The ratio of poly (acrylate) concentration in the
copolymer to the initial feed concentration was calculated
and is also shown in Table 4.5. This ratio can be seen as a

131

b c c b

f-C-CH ? CH 2 -CH 2 .CH 2 -C— NH

H 2 S0 4

c,c

b,b'

n Uww>ywM* fi^**»

11 10 9 8 7 6 5 4 3 2 1 ppm

Figure 4.42 ^-NMR spectra of PhDAA homopolyamide .

132

-f-C-CH?Cht-CH2CH 2 -C— NH

*' r J I 9

H 2 S0 4

Nh+)— (-C-CH2-NH+);

■ w j

S< » i »w M H aj M^i

CH2

CH2 f

CH3-CH2CH2-CH2-0 — C-CH e
abed vU

N i Mai * i»

1110 987 65 4 32 10 ppm

Figure 4.43 ^-NMR spectra of 33PhDAA-g-66BAx

133

h H H h

O O \ / O

-(-C-CH2CH2-CH2CH2-C-NH— f ^ NH-)— fc-CH 2 -NH-)-

;j ;>

1 J J 1

H 2 S0 4

CH2

h H H h

CH 2 f

CH3-ChbCH2CH20— C-CH e
abed \L/ a

^' ' ' '1 nrii J i- -im*~

' I I I I I I I II I I P I II I I I I I I I I I I J I I I I I I I I l| I II I I I I I I I I 1 l| MM |l I I

11 10 9 8 7 6 5 4 3 2 1 ppm

Figure 4.44 ^-NMR spectra of 10PhDAA-90BAx.

134

O O \ ( o

-f-C-CH 2 -CH 2 -CH 2 .CH 2 -C— NH
g' h' h g

NH-)— (-C-CH 2 -NH-)-

x y

CH 2

Figure 4.45 1 H-NMR spectra of 33PhDAA-66FAx ,

135

H 2 S0 4

r^%

o
nh4— (-c-ch 2 -nh4-

CH 2

S
CH, f

CH o'CH p'CHp'CHp'O — C — CH g

abed 4^

i ' ' ' ' i ' ' ' ' i ' ' ' ' i

11 10 987 65 4 3 210 ppm

Figure 4.46 ^-NMR spectra of 66PABA-33BAx,

136

H 2 S0 4

y J NH^-fC-CH 2 -NH^-

CH 2

S
CH 2 f

CH 3 -CH 2 -CH 2 CH 2 — C-CH e

abed 4^

O zT

11 10 9 8 7 6 5 4 3 2 1 ppm

Figure 4.47 X H-NMR spectra of 33PABA-66BAx .

137

TABLE 4.5 Chemical composition of purified graft copolymers
from elemental analysis and NMR.

Sample ID

Nitrogen

Graft Copolymer

[Acrylate] copolymer**

(wt. %) EA

Compos ition EA
(wt. %)*

Amide Acrylate

Composition™"
(wt. %)

Amide Acrylate

[Acrylate] feed

66PABA-g-33BAx

9.47

81 19

79 21

0.61

33PABA-g-66BAx

6.35

54 46

60 40

0.65

33PhDAA-g-66BAx

7.03

55 45

54 46

0.69

10PhDAA-g-90BAx

6.42

46 54

47 53

0.60

33PhDAA-g-66FAx

9.64

75 25

0.38

Calculated using wt %N for PABA=11.76, PhDAA=12.84.
** based upon average of EA and NMR acrylate wt . %.

measure of % poly (acrylate) incorporation. The values for
the poly (butyl acrylate) containing graft copolymers,
between 0.60 to 0.69, do not vary greatly. In other words
the p(BA) content in the purified graft copolymers is
approximately 60 to 69% of that in the feed. The result for
the fluoroacrylate graft copolymer differs significantly
with a copolymer to feed ratio of 0.38. This decrease in
macromonomer incorporation can again be attributed to the
decrease in macromonomer functionality from inefficient
chain transfer for the fluoroacrylate polymerization.

Inherent viscosities of graft copolymers . The
intrinsic viscosities of the polyamide graft copolymers were
measured in concentrated sulfuric acid at a concentration of
0.005g/ml. The results are shown in Table 4.6.

138

TABLE 4.6 Inherent viscosities of polyamide graft
copolymers .

Sample ID

Graft Copolymer
Composition*
(wt %)

Amide Acrylate

Tlinh**

PABA

100

1.16

66PABA-g-33BAx

80 20

0.88

33PABA-g-66BAx

57 43

0.55

PhDAA

100

1.24

33PhDAA-g-66BAx

55 45

0.52

10PhDAA-g-90BAx

47 53

0.37

33PhDAA-g-66FAa

75 25

1.11

♦average of EA and NMR results

**measured in cone. H 2 S0 4 at 30°C, 0.005g/ml

The Tjinh values of the PABA and PhDAA homopolyamides are
1.16 and 1.24 respectively. The viscosity decreases as the
poly (butyl acrylate) content in the graft copolymer
increases. This decrease in viscosity is not as dramatic
for the 25% FA grafted polyamide whose r|i n h is similar to
that of the PhDAA homopolyamide, 1.11 vs. 1.24 respectively.
By comparison, only 20% BA causes a decrease in the r|i nh of
PABA from 1.16 to 0.88.

These viscosity values can be used as an indication of
the molar masses of the graft copolyamides . Quantitative
molar mass values cannot be derived since the precise
relationship between molar mass and viscosity is not known
for these copolymers. However, it has been shown in the

139

copolymerization of other amino acids (Hig80b) that the
viscosities determined for polyamides synthesized in
Higashi's solvent system translate into molar mass values of
approximately 15 to 35kg/mol.

4. 2. 3 .Blends of Graft Copolymers with Nylon 6

Preliminary experiments were performed in order to
study the effect of graft copolymer addition on the
mechanical properties of Nylon 6. BA-containing graft
copolymers were blended with Nylon 6. Graft copolymers were
also added to Nylon 6/Poly (butyl acrylate) incompatible
blends in order to investigate the ability of the graft
copolymer to act as a surfactant or compatibilizer .

The synthesized copolymers, PhDAA, 33PhDAA-g-66BAx, and
10PhDAA-g-90BAx were blended with Nylon 6. Also, 33PhDAA-g-
66BAx and 10PhDAA-g-90BAx were added to an 85/15 blend of
Nylon 6 and 65kg/mol poly (butyl acrylate). Blend and sample
preparation has been described previously (section 3.2.2).
Tensile properties of all blends can be found in Table 4.7.

140

TABLE 4.7 Tensile properties of nylon 6 blends

Material

Modulus
(GPa)

Proportional Limit

Stress
(MPa)

Elongation
(%)

Stress at

Failure

(MPa)

Elongation @

Failure
(%)

Nylon 6

90 N6/10
PhDAAx

85 N6/15

33PhDAA-g-

66BAx

85 N6/15

lOPhDAA-g-

90BAx

1.60± 0.07 16.2+2.0 1.0± 0.2

2.85+0.46 27.4+4.6 0.98+0.08

1.02+0.03 9.95+1.03 1.0+0.1

0.91+ 0.09 10.81+ 1.03 1.0+ 0.2

90 N6/10 PBA 1.32+0.21 15.0+1.6 1.2+0.3

90 N6/10 PBA 1.32+ 0.05 19.2+ 2.4 1.3+ 0.1
w/0.6 wt%
33PhDAA-g-
RA66x

90 N6/10 PBA 1.38+0.05 18.2+1.5 1.5+0.2
w/0.6 wt%
lOPhDAA-g-
90BAx

71.6+ 5.1
83.0+ 7.3

28.0+ 3.0

24.4+ 4.1

79.8+ 9.8
63.9+ 4.5

68.0+ 7.2

55+9
8.0+ 1.1

7.2+ 1.9

8.4+ 1.9

50+ 29
22+8

20+ 14

Addition of PhDAA to Nylon 6 produced an increase in
modulus and a severe decrease in ductility. The modulus
increased from 1.60 to 2.84 GPa with the addition of 10%
PhDAA but the percent elongation at failure decreased from
55 to 8%. The proportional stress increased from 16.2 to
27.4 MPa upon PhDAA addition. The aromatic polyamide
domains behave like hard inclusions within the nylon matrix,
reinforcing nylon and serving as sites for stress
concentration. This behavior is partly due to the visibly
poor mixing between the nylon and the aromatic polyamide or
graft copolymer. Dispersion is poor as off-colored regions
of brown PhDAA are clearly visible.

141

Poor mixing is a result of the lack of thermal
transitions for the amide phase not allowing flow during
compression molding. The TG/DTA (Figure 4.48) and DSC
(Figure 4.49) traces of the 33PhDAA-g-66BA graft copolymer
and PhDAA display no thermal transitions prior to
degradation for the amide phase. The only thermal
transition observed is the glass transition of p(BA) .
Yoshimitsu (Yos94) has reported that similar aromatic
polyamides degrade before their melt temperature.

As expected, the graft copolymer blend with Nylon 6
also exhibits poor mixing due to the lack of mobility or
flow during compression molding. Regions of graft copolymer
are clearly visible within the nylon 6 film. However, the
modulus, proportional stress , stress at break, and
elongation at break of the blend containing 15% graft
copolymer are significantly lower than that of the either
the neat nylon 6 and the PhDAA/Nylon 6 blend. The large
domains of graft copolymer do not reinforce the blend due to
the presence of the rubbery p(BA) segments. In essence, we
have gone from a system with hard inclusions to one with
soft inclusions. These inclusions are very poorly dispersed
and thus no toughening of the system is observed.

The differences between the mechanical properties of
the 85/15 blends of nylon 6/33PhDaa-g-66BAx and nylon
6/10PhDAA-g-90BAx are not significant. The final
compositions of the graft copolymers are not as dissimilar
as their feed compositions. From Table 4.5, we have already

142

r
o

Figure 4.41

33PhDAA-g-66BA

PhDAA

r- 180
160
140
120

- 100

- 80

- 20

175 350 525 700

Temperature (°C)

825

TG/DTA analysis of 33PhDAA-g-66BAx graft
copolymer and PhDAA homopolyamide .

- 40

143

1000

500 -

-500 -

-1000

-1500

-2000 H

-2500

-45°C
P(BA) Tg

PhDAA

33PhDAA-g-66BA

100 200

Temperature (°C)

300

Figure 4.49 DSC analysis of 33PhDAA-g-66BAx graft copolymer
and PhDAA homopolyamide .

144

shown that these two graft copolymers contain 45 and 55% BA
respectively.

The graft copolymer had little effect as a surfactant
in the Nylon 6/poly (butyl acrylate) blend. Again, lack of
mixing would prevent the graft copolymers from migrating to
the interface between the immiscible polymers. Also, the
poly (butyl acrylate) homopolymer migrated to the surface of
the blend and thus the mechanical properties of these blends
may not be representative of a 10% p(BA) content. The films
had to be wiped with acetone prior to tensile testing in
order to avoid slippage.

4.3. Offsetting Polymerization Shrinkage in Dental Resins
through the Incorporation of Maleic Anhydride

Volumetric shrinkage is a major problem inhibiting the
long term success of current dental restorative materials
(Bau82). This shrinkage is inherent to the free radical
polymerization of the multifunctional methacrylate resins
used in the dental composites. The volumetric shrinkage is
due to the reduction in molar volume, or spacing between
monomer units, that occurs when vinyl compounds are
polymerized. Polymerization shrinkage creates both a weak
interface between the tooth structure and the restoration as
well as residual stresses within the composite structure,
leading to premature failure in the restoration (Bra86) .

BisGMA is the predominant monomer currently in use in
dental composites. Although the shrinkage is decreased from

145

that of simple methacrylates, BisGMA based composites still
exhibit volumetric shrinkage values of around 3% (Sul93) .
There have been some modifications to BisGMA. However, the
differences are not generally significant and many
manufacturers still rely on the BisGMA monomer for their
dental composites.

There are many research programs focused on different
chemical structures and processes that will reduce
polymerization shrinkage (Bra92, Bye92, Sta91). The main
thrust in dentistry has been the spiro orthocarbonate based
systems which are the result of early pioneering work by
Bailey (Bai72). The spiro orthocarbonates involve a dual
ring opening mechanism which increases the molar volume of
the polymer compared to that of the monomer. The deficiency
with these ring opening systems is the cost of the reactive
monomer and the slow kinetics. Typically dental restorations
can be cured within a few minutes whereas the spiro
orthocarbonates are very slow reacting (Bra92, Bye92) .

Although not ideal, these systems do provide some
insight as to a logical step in the evolution of dental
restoratives. The ring opening polymerization kinetics may
be too slow, yet the ring opening mechanism does provide a
net increase in molar volume. Thus, by combining the fast
kinetics of the methacrylate resin with the ring opening
reactions, one may reduce the polymerization shrinkage.

The focus of our work is to incorporate cyclic
anhydride functionality into dimethacrylate networks. When

146

maleic anhydride (MA) is ring opened by hydrolysis to maleic
acid there is a corresponding theoretical 10% increase in
molar volume. By incorporating maleic anhydride into common
dental resins, such as a BisGMA/TEGDMA system, and then
hydrolyzing the anhydride, we should offset some of the
shrinkage associated with polymerization.

4 . 3 . 1 . Copolymer Compositions

A design study including the composition matrix shown
in Table 4.8 was carried out. The composition ranges are 50-
70 wt . % triethyleneglycol dimethacrylate, 20-40%
propoxylated BisGMA (pBisGMA) , and 0-20% maleic anhydride.
Propoxylated BisGMA was substituted for BisGMA in order to
avoid any reaction of the anhydride with the pendant
hydroxyl group in BisGMA. Monomer structures are shown in
Figure 4.49. All samples, about 2mm thick, were cured
between glass plates using 0.4 wt . % AIBN at 75°C for 12
hours followed by a postcure at 160°C for 2 hours.

Although dental composites are generally light cured in
actual application, a heat cure was used to provide an ideal
structure with maximum conversion. Maximum conversion is
desired in order eliminate this as a variable in the
analysis of property differences between anhydride and non
anhydride samples and thus to prove the feasibility of this
approach. Also, due to the differing reactivity ratios
common in maleic anhydride-methacrylate copolymerizations

147

Propoxylated BisGMA
/ =v CH 3 O

CH, X / CH, N ' C

Triethyleneglycol dimethacrylate
o o

CH 3 — CH -C -O — (CHj-CHj-O)^— C — CH -CH 3
CHj CH2

2-phenylethyl methacrylate Maleic anhydride

CH 3 CH=CH

CH 2 =C J \

C=0 O^ ">W ^O

o
I

CH 2
I
CH 2

"\^

Figure 4.50 Chemical structures of methacrylate and

anhydride monomers for dental applications

148

TABLE 4.8 Experimental matrix of dental monomer
compositions .

Sample ID Weight % in Monomer Mixture

TEGDMA pBisGMA MA

DM1 70 30
DM2 70 20 10
DM3 60 40
DM4 60 30 10
DM5 60 20 20
DM6 50 40 10
DM7 50 30 20

(Bra89) , high conversions are desired in order to maximize
anhydride incorporation.

Polymerizations of 2-phenylethyl methacrylate (PEMA)
with maleic anhydrides were used to model the dimethacrylate
reaction. They were synthesized in bulk using 0.4 wt% AIBN
at 75°C for 4 hours. The resulting product was dissolved in
chloroform, precipitated in ether to remove unreacted
monomer. The polymer was then filtered and dried under
vacuum at 40°C overnight.

We are using PEMA as a linear analog of the
dimethacrylate BisGMA type monomers. PEMA copolymers with
maleic anhydride should be linear and therefore soluble in
common organic solvents. Therefore, we will be able to
monitor the effect of anhydride incorporation on properties

149

such as molar mass and glass transition temperature which
are more difficult to analyze in highly crosslinked samples.
PEMA copolymer compositions are listed in Table 4.9.

TABLE 4.9 PEMA-maleic anhydride monomer compositions.

Sample ID Weight % in Monomer Mixture

2-phenylethyl maleic anhydride

methacrylate

PEMA1 100

PEMA2 90 10

PEMA3 8 2

PEMA4 7 30

PEMA5 60 40

4 . 3 . 2 .Copolymer Characterization

Dimethacrylate dental resins . Equilibrium water
content (EWC) was measured gravimetrically and the
extraction of unreacted anhydride was monitored using UV
spectroscopy of H 2 solutions from swelling experiments.
Swollen, equilibrated samples were then dried under vacuum
at 60°C to constant weight. Density measurements of the
original dry samples and the post-hydrolysis dried samples
were calculated using Archimedes' principle. These density
changes along with changes in sample mass were used in order
to calculate volume changes from hydrolysis. Details of the

150

experimental methods and subsequent calculations described
above have been previously discussed in section 3.2.3. The
results of this study are shown in Tables 4.10. and 4.11.

TABLE 4.10 EWC of maleic anhydride dental resins.

Sample Composition Equilibrium H 2 Content

(TEGDMA/pBisGMA/MA) [%)

70/30/0 4.12
70/20/10 7.43
60/40/0 3.81
60/30/10 6.51
60/20/20 11.53
50/40/10 6.08
50/30/20 10.94

The dominant factor affecting the water uptake, shown
in Table 4.10, is the anhydride content of the copolymer.
Values vary from 4% for the two samples with no anhydride to
11% for the two samples containing 20% anhydride. This
result is expected considering the hygroscopic nature of the
anhydride functionality and its susceptibility to hydrolysis
to a diacid. An increase in TEGDMA relative to pBisGMA also
seems to increase the EWC, but to a much lower extent than
maleic anhydride.

These swollen equilibrated samples were thoroughly
dried back to a constant weight. Anhydride containing

151

samples showed a residual weight gain up to 1.55% as shown
in Table 4.11. This weight gain is believed to be due to
hydrolysis of the anhydride functionality. Multiple
processes are occurring during the water uptake. Not only
is the hydrolysis of the anhydride driving further water
uptake, but unreacted maleic anhydride is being converted to
maleic acid and then extracted from the sample.

TABLE 4.11 Residual weight gain, anhydride incorporation
and post polymerization expansion of maleic
anhydride dental resins.

Sample % Efficiency of Post

Composition residual Anhydride polymerizati
TEGDMA/pBisGMA/MA weight Incorporation* on % volume

gain

expansion**

70/30/0

0.00

70/20/10

1.10

0.94

1.07

60/40/0

0.04

0.00

60/30/10

1.11

0.94

1.08

60/20/20

-0.04

0.84

50/40/10

1.53

0.97

1.67

50/30/20

1.55

0.91

2.11

*Determined from the UV analysis of H 2 extraction solutions
**Volume calculated from mass and density measurements

UV spectroscopy was run on the swelling solutions. A

calibration curve for the UV absorbance of maleic acid was

determined in order to measure the maleic acid concentration

in the extractables . All anhydride not extracted was

152

assumed to be incorporated. Incorporation efficiencies
ranged from 0.84 to 0.97.

Post polymerization volume expansions, measured using
mass and density values, of up to 2% were calculated for
anhydride containing samples. The expansion increases with
increasing anhydride concentration.

TG/DTA shows a significant difference in decomposition
for anhydride containing polymers as seen in Figure 4.51.
Initial weight loss occurs at a lower temperature than the
pure methacrylate resins but the anhydride increases the
thermal stability at higher temperatures. This is
consistent with reported results of MMA-maleic anhydride
copolymers (Bhu88). Increased thermal stability may be due
to the anhydride unit blocking the 'unzipping'
depolymerization reaction common in poly (methacrylates) .

Linear poly (PEMA-co-MA) copolymer analogs . FTIR
spectra were collected to verify the presence of anhydride
in the structure. Figure 4.52 shows the comparison of the
carbonyl region for poly (phenylethyl methacrylate) and a
copolymer with 40 wt . % maleic anhydride in the feed. The
absorption at 1727cm" 1 can be assigned to the C=0 stretch
from the ester carbonyl in the methacrylate. The absorption
of the carbonyl for the anhydride is present as a doublet at
1781 and 1853cm -1 , for the asymmetric and symmetric C=0
stretch .

153

100

i 1 r—

1 1 1 1

■"••^v

l «s

- 60/40 TEGDMA/pBisGMA
-50/30/20

TEGDMA/pBisGMA/MA -

\
v \

1 1 !

.11,-

100 150 200 250 300 350 400 450 500 550 600

Temperature (°C)

Figure 4.51 TG/DTA of anhydride modified dental resin,

154

O
CO

-Q

i—

c/)

-Q

poly(60PEMA-co-40MA)

1728

1800 1600

Wavenumber (cm-i)

Figure 4.52 FTIR spectra of poly(PEMA) and poly ( 60PEMA-CO-
40maleic anhydride) .

155

The relative intensities of the 1726cm" 1 and 1781cm" 1
carbonyl absorptions were used to calculate the ratio of
PEMA to anhydride in the final copolymers. The final
copolymer anhydride content, along with the measured glass
transition temperatures, are shown in Table 4.12. The glass
transition temperature of the PEMA copolymers increases with
anhydride content from 43.8°C for neat poly (phenylethyl
methacrylate) to 63.2°C for the 60/40 PEMA/MA copolymers.
Actually, the anhydride content in the feed was 40 wt . % but
the anhydride content in the copolymer is closer to 22%.

TABLE 4.12 Glass transition temperatures and composition of
PEMA-maleic anhydride copolymers.

Polymer Tg* Anhydride

(°C) content**

(wt.%)

Poly(PEMA) 43.8

Poly (90PEMA-co-10MA) 47.8

Poly (80PEMA-CO-20MA) 54.7 12

Poly (70PEMA-CO-30MA) 57.9 14

Poly(60PEMA-co-4 0MA) 63.2 22

^Measured by DSC.

** Calculated from the relative intensities of the 1726cm" 1 and 1781cm" 1

carbonyl absorptions using the monomer mixture as a standard.

156

Efficiency of anhydride incorporation was significantly
lower than in the dimethacrylate copolymerizations . The
anhydride content in the PEMA copolymers is approximately
50% of that in the feed. Because of the difference in
reactivity ratios in maleic anhydride-methacrylate
copolymerizations, high conversions are required to produce
polymer high in anhydride content. Conversion, and
therefore anhydride incorporation, in this system is
expected to be lower than that in the dimethacrylate system
due to the lower cure temperature and shorter cure time.

The effect of anhydride feed concentration on the molar
mass averages of the copolymers is illustrated in Table
4.13. Increasing the anhydride content leads to a severe
reduction in the average molar mass of the resulting
copolymer. The Mn decreases from 63kg/mol for neat PEMA to
26kg/mol for the copolymer with a 60/40 PEMA/MA feed ratio.

157

TABLE 4.13 Molar mass averages from GPC for PEMA-anhydride
copolymers .

Polymer Mn Mw PDI

( kg/mol ) ( kg/mol )

Poly(PEMA) 62.9 245.5 3.9

Poly (90PEMA-CO-10MA) 44.5 249.8 5.6

Poly (80PEMA-CO-20MA) 30.5 192.7 6.3

Poly(70PEMA-co-30MA) 36.0 169.3 4.7

Poly(60PEMA-co-40MA) 25.9 122.2 4.9

CHAPTER 5
SUMMARY AND CONCLUSIONS

The synthesis and characterization of amino acid
terminated poly (acrylate) macromonomers using a cysteine
chain transfer agent has been studied. The ability of these
macromonomers to react with amide precursors in a
condensation graft copolymerizations has been investigated.
Although the novelty of this approach in terms of
macromonomer synthesis, chain transfer chemistry, and
condensation graft copolymerizations will be discussed, it
is important to note the motivation for this study.

Although at first glance the macromonomer and graft
copolymer work does not seem to be closely related to our
studies of dental resins, the motivations for these studies
are tied closely together. The efforts of this research
group have been focused on one main goal- increasing the
understanding of the role of interfaces in composites and
multiphase copolymer systems.

The interfacial bonding between two phases is critical
in determining the overall properties of composite (Arn97) .
Much work has concentrated on surface modification in fiber
and particulate reinforced composites (Arn97, Ore97) as well
as compatibilizing agents in polymer-polymer composites with

158

159

the goal of reducing interfacial tension and increasing
interfacial bonding.

One of the composite systems we have been investigating
is the glass particulate reinforced poly (dimethacrylate)
commonly used as dental restorative materials. The failure
of these dental composites, and poor lifetime performance as
compared to amalgam restorations, is generally attributed to
a poor interface (Bra86). Although the glass-resin
interface has been studied extensively, the source of
failure is usually the interface between the composite
restoration and the remaining tooth structure (Sod91) .

Two of the main sources of this poor interface are as
follows :

1) the polymerization shrinkage during composite cure
causing the restoration to pull away from the remaining
tooth structure. This leads to marginal leakage, the
infiltration of saliva and bacteria under the
restoration, which can lead to the reincidence of
caries .

2) poor bonding between the exposed tooth structure,
composed of hydrophilic proteinaceous dentin tubules,
and the hydrophobic dimethacrylate composite.

It is to this effect that a large extent of the research in
dental composites has focused on one of two areas, the
reduction or elimination of polymerization shrinkage or the
evaluation of new dentin bonding agents. Only our study on
offsetting polymerization shrinkage through the addition of

160

maleic anhydride deals specifically with one of these
problems .

However, the macromonomer and graft copolymer work was
targeted to determine the ability to synthesize a copolymer
capable of interacting with both the hydrophobic
methacrylate and hydrophilic dentin structure at the tooth-
restoration interface. Although the synthesized amide-
acrylate graft copolymers were not tested in such a system
and their aromatic structures may not make them suitable for
this application, the feasibility of synthesizing the
desired structures has been shown. The groundwork has been
laid to synthesize acrylate graft copolymers with other
amino acids more suitable to this particular application.

Although preliminary work performed here (section
4.2.3) on the possible applications of the synthesized graft
copolymers was inconclusive, these graft copolymers may be
useful in other interfacial applications such as
compatibilizing agents in immiscible polymer blends, rubber
modifiers, and surface modifiers.

5.1. Chain Transfer Functionalization of Poly (acrylates) and

Poly(methacrylates)

Studies on macromonomer synthesis have concentrated on
that of vinyl f unctionalized macromolecules . These
macromonomers are thus capable of copolymerizing with other
vinyl monomers in the synthesis of addition-addition graft
copolymers. The novelty of the approach to macromonomer

161

synthesis applied herein lies in the ability to synthesize
free radically polymerized addition poly (acrylates ) capable
of participating in a condensation graft copolymerization
with polyamides. It is the amino acid functionality which
makes these macromonomers ideal for grafting or binding to
polyamides or proteins. The conclusions for the chain
transfer f unctionalization reactions are summarized below.

5 . 1 . 1 .Conclusions for Preliminary Evaluation of Cysteine
Chain Transfer Agent

1) Under neutral pH conditions, the chain transfer

f unctionalization of poly (butyl acrylate) using a
cysteine chain transfer agent is ineffective. This
reaction produces a polymer with a broad molar mass
distribution, PDI= 4.8, and a higher than expected Mn.

2) Chain transfer is inefficient due to consumption of
cysteine in an undesired side reaction. Under neutral
conditions, 5% of cysteine is ionized. The sulfur
anion can react with butyl acrylate to form S-
carbobutoxyethylcysteine . The consumption of cysteine
by this side reaction leads to a decrease in effective
cysteine concentration over time and therefore, reduced
termination by chain transfer.

3) The formation of the undesired side product can be
virtually eliminated by reducing the pH of the
reaction. A severe reduction in molar mass from

162

effective chain transfer is observed upon acidification
of this reaction.

5 . 1 . 2 . Conclusions for Synthesis and Characterization of
Amino Acid-terminated Poly (butyl acrylate)

1) The chain transfer polymerization of butyl acrylate in
the presence of cysteine is highly effective. The
calculated chain transfer constant, 1.49, is higher
than that of other common chain transfer agents.

2) The presence of an amino acid functionality from the
cysteine end group is verified by a combination of FTIR
spectroscopy, NMR, ICP, as well as by the ability of
the macromonomer to react with amide precursors in a
condensation polymerization.

3) Predicted functionalities, using the Mayo model, are
higher than those measured by the ICP-GPC technigue.
This could be due to non-ideal conditions such as chain
transfer to solvent, high conversions, etc.. and could
also be affected by the inherent error in functionality
determination using GPC molar mass values.

5 . 1 . 3 . Conclusions for Synthesis and Characterization of
Amino Acid-terminated Poly (MMA-co-QFPMA)

Cysteine is not as effective, with a chain transfer
constant of 0.26, in the chain transfer
copolymerization of methyl methacrylate and
octaf luoropentyl methacrylate. At equivalent levels of

163

cysteine, much higher molar masses are observed in this
reaction than in the p(BA) polymerization.

2) The difference between these two systems may be
explained by the difference between acrylates and
methacrylates . Other research groups (NAIR) have
observed, although not to such a large extent, lower
chain transfer constants for methacrylates. This has
been attributed to the lower reactivity of
methacrylates, leading to degradative chain transfer.

3) The lower chain transfer constant and subsequent higher
molar mass leads to lower predicted functionalities
when compared to p(BA) macromonomers . Functionality
could not be determined experimentally due to the high
molar mass and consequent low end group concentration
of the f luoropolymer .

5.2. Graft Copolymerizations of Macromonomers with

Polyamide Precursors

There are few reports of well defined polyamide graft
copolymers with addition polymers such as poly (acrylates) or
poly (methacrylates) . Most studies of amide graft
copolymerizations with addition polymers involve either the
in situ formation of graft copolymers in polymer blends or
radiation induced surface graft techniques. Although
effective for their intended applications, neither method
produces a well defined graft copolymer that can be isolated
and studied.

164

The objective of this study was to synthesize such
graft copolymers using amino acid terminated macromonomers
synthesized previously. A summary of the conclusions for
the graft copolymerization of these macromonomers is listed
below.

5 . 2 . 1 . Conclusions for Synthesis of Poly (amide-g-acrylate!
Graft Copolymers

1) P(BA) and p(FA) were polymerized with aromatic and
aromatic-aliphatic amide precursors in a P(0Ph)3
catalyzed polycondensation reaction.

2) The products were extracted with THF and subsequent GPC
analysis on extractables showed a bimodal distribution.

3) GPC-UV results indicate that both components in the
extractables contain aromatic segments from the
polyamide. This would indicate that the THF soluble
fraction is not composed of purely of unreacted or
unfunctionalized poly (acrylates) , but contains either
low molar mass graft copolymer or poly (acrylate) rich
graft copolymer.

4) Weight loss from extraction was much higher for p(FA)
graft copolymers than for equivalent p(BA)
compositions. This can be explained by an increase in
unfunctionalized chains as predicted by the lower chain
transfer constant in the macromonomer polymerization.

165

5 . 2 . 2 . Conclusions for Characterization of Poly (amide-g-
acrylate) Graft Copolymers

1) The presence of both amide and acrylate segments in the
graft copolymer is verified by FTIR and NMR.

2) Acrylate content was determined both by NMR and
Elemental Analysis. Acrylate concentration in the
copolymer is proportional to that in the feed and
varied from 25 to 55 wt . % .

3) For p(BA) graft copolymers, the concentration of p(BA)
in the copolymer is 60-69% of the p(BA) in the feed.
This is due to both the presence of unfunctionalized
chains as well as the solubility of low molar mass,
high p(BA) graft copolymer in the THF extraction.

4) The incorporation efficiency for the p(FA) is much
lower, 38%, due to the lower predicted functionality of
these macromonomers . Steric effects from the higher
molar mass of the macromonomer may also be limiting
their incorporation into high molar mass graft
copolymer .

5) Inherent viscosities of graft copolymers are consistent
with those of polyamides with molar mass values around
15 to 35kg/mol (Hig80b) . The graft copolymer viscosity
decreases with increasing p(BA) concentration. The
effect is less dramatic for graft copolymers containing
the higher molar mass f louroacrylate macromonomer.

166

5 . 2 . 3 .Conclusions for Mechanical Properties of Nylon 6/Graft
Copolymer Blends

1) Addition of PhDAA to Nylon 6 produces an increase in
modulus and a severe decrease in ductility. The graft
copolymer domains are behaving like hard inclusions
within the nylon matrix.

2) This behavior is due to the visibly poor mixing between
the nylon and the PhDAA. Poor mixing is a result of
the lack of thermal transitions for the amide phase on
the graft copolymer not allowing flow during
compression molding.

3) Blending of the graft copolymer caused a decrease in
all mechanical properties. Again mixing was extremely
poor and domains were clearly visible.

4) The graft copolymer had little effect as a surfactant
in the Nylon 6/poly (butyl acrylate) blend. Again, lack
of mixing would prevent the graft copolymers from
migrating to the interface between the immiscible
polymers .

5.3. Offsetting Polymerization Shrinkage in
Poly (dimethacrylate) Dental Resins

Volumetric shrinkage is a major problem inhibiting the
long term success of current dental restorative materials.
This shrinkage is inherent to the free radical
polymerization of the multifunctional methacrylate resins
used in the dental composites. Polymerization shrinkage

167

creates both a weak interface between the tooth structure
and the restoration as well as residual stresses within the
composite structure, leading to premature failure in the
restoration. Although various methods and different
monomers have been studied to alleviate this problem, no
solution to date has been discovered that eliminates
shrinkage without significantly altering the cure and
mechanical properties of the resin.

The goal of this study was to demonstrate that we can
offset the polymerization shrinkage of common dimethacrylate
resins without significantly changing the comonomer
structures through the addition of maleic anhydride. The
conclusions for this work are summarized below.

5. 3 . 1 .Conclusions for Characterization of Maleic Anhydride-
containing Dental Resins

Maleic Anhydride incorporation into heat cured dental
resins was verified by FTIR, extraction, and UV
spectroscopy. Efficiency of incorporation is between
0.84 and 0.97. High levels of incorporation are due to
high levels of conversion expected for high cure
temperatures and long cure times.
The equilibrium water content of the resins is
proportional to the concentration of maleic anhydride,
varying from 3 to 10% H 2 0. This can be explained by
the hygroscopic nature of anhydride functionalities and
resulting diacid functionalities.

168

Mass and density values of dry samples, before and
after extraction, were measured in order to determine
volume changes. Volume expansions of up to 2% were
measured for anhydride-containing samples.

5 . 3 . 2 .Conclusions for Synthesis and Characterization of
Anhydride Copolymer with PEMA

Maleic anhydride copolymers with 2-phenylethyl
methacrylate (PEMA) were synthesized in order to
determine the effect of the anhydride on properties
more readily measured in non crosslinked polymers.
Efficiency of anhydride incorporation was significantly
lower than in the dimethacrylate copolymerizations,
ranging from 0.45 to 0.60. Because of the difference in
reactivity ratios in maleic anhydride-methacrylate
copolymerizations, high conversions are required to
produce polymer high in anhydride content. Conversion,
and therefore anhydride incorporation, in this system
is expected to be lower than that in the dimethacrylate
system due to the lower cure temperature and shorter
cure time.

Increasing the anhydride content leads to a reduction
in average molar mass yet a higher glass transition
temperature in the resulting copolymer.

CHAPTER 6
FUTURE WORK

6.1. Macromonomers and Graft Copolymers

The overall results of this work indicate that graft
copolymers of polyamides and poly (acrylates) can be
successfully synthesized using cysteine end capped
macromonomers. Although a series of macromonomers and graft
copolymers was synthesized and characterized, optimization
of the individual steps and extension of this work to other
reaction systems should be investigated. Suggested further
experiments are outlined below.

6 . 1 . 1 .Macromonomer Work

1) The effect of further reducing AIBN content on
functionalization efficiency should be determined. As
was stated previously, any chains initiated by AIBN
will be unfunctionalized and reduction of AIBN
concentration should yield more highly f unctionalized
macromonomers. However, this is expected to be
accompanied by a drop in overall conversion.

2) The macromonomer synthesis could be extended to other
monomers. For example, crosslinked systems may be
investigated. It is expected that amino acid

169

170

functionalized networks could be synthesized by the
polymerization of dimethacrylates or any divinyl
monomer in the presence of cysteine.

6. 1.2. Graft Copolymers

1) The reaction of acrylate macromonomers with a more
processable polyamide would be of interest. Those
studied here were soluble only in strong acids and
showed no distinct thermal transition prior to
degradation.

2) A more soluble polyamide would allow for solution
processing either for blending with other polymers or
for more precise graft copolymer molar mass
determination (GPC, light scattering) .

3) More importantly, a polyamide that is thermally
processable is of interest. Those synthesized in this
study, similarly to other aromatic polyamides, degrade
before they melt thereby complicating melt processing.
Melt processability would allow for blending with other
polymers and it would facilitate the sample preparation
for the mechanical testing of the graft copolymers.

4) It has been reported that, although these aromatic
polyamides generally do not display a distinct melt
prior to degradation, they are semicrystalline (Yos94) .
X-ray diffraction on the synthesized graft copolymers

171

would be useful to determine their level of
crystal Unity.

5) If successful, the effect of macromonomer molar mass and
content on the crystallization and mechanical properties
of the graft copolymers would be of interest.

6) The graft copolymerization of acrylates with naturally
occurring amino acids such may be of interest in
biological applications. Higashi (Hig80b) has already
shown that poly (leucine) , poly (valine) , poly (alanine)
and poly (phenylalanine) could be synthesized in the NMP
solvent system used in this study. In fact, high molar
mass poly (amino acids) up to 35kg/mol were synthesized.
The polymerization with amino acids is where cysteine
terminated polymers will be especially beneficial
because of the built in stoichiometry of the amino acid
functionality.

7) Other methods of copolymerization of these macromonomers
should be investigated. Interfacial polymerizations or
bulk polymerization with caprolactam for nylon 6
copolymers may be possible.

6.2. Anhydride-containing Dental Resins

Further studies in this research group on anhydride
containing dental resins has been ongoing (Gra95, Wes97,
Sch97, Meh97) . The work presented here represents the
initial foundation for the ongoing studies. The effect of

172

anhydride incorporation on mechanical properties has been
studied in detail (Gra95) . The incorporation and modeling
(Wes97) of other anhydrides including nadic methyl anhydride
(Meh97) and methacrylic anhydride (Sch97) has been
investigated.

The light or room temperature cure polymerizations of
these copolymers has presented unique problems.
Incorporation of anhydride is not extensive as compared to
the heat cure systems (Gra95) . However several methods of
improving the incorporation into dental resins- either
through the addition of prepolymerized anhydride copolymer
or through the addition of monomers more reactive with
maleic anhydride, such as styrene or vinyl ethers- are
currently under evaluation.

LIST OF REFERENCES

Alb86 Albrecht, K. and Wunderlich, W. Agnew. Makromol .
Chem. , 145/146, 89 (1986) .

Ana89 Anastasiadis, S.H.; Gancarz, I. and Koberstein,
J.T. Macromolecules, 22, 1449 (1989).

Ath77 Athey Jr., R.D. and Mosher, W.A. J. Polym. Sci.
Chem. , 15, 1423 (1977) .

Arn97 Arnold, J.J. Ph.D. Dissertation, University of
Florida (1997) .

Bai72 Bailey, W.J. and Sun, R.L. Am. Chem. Soc . , Div.
Polym. Chem., Polym. Prepr . , 13, 281 (1972).

Bak65 Baker, W.P. Jr. U.S. Patent 3,225,119 (1965).

Bar94 Barbosa, R.V.; Soares, B.G. and Gomes, A.S. J.
Appl. Polym. Sci., 47, 1411 (19 93).

Bau82 Bausch, J.R.; De Lange, C . ; Davidson, C.L.;

Peters, A. and De Gee, A.J. J. Prosthet. Dent.,
48, 59 (1982) .

Bec96 Beck Tan, N.C.; Tai, S.K. and Briber, R.M.
Polymer, 37, 3509 (1996) .

Bhu88 Bhuyan, K. and Dass, N.N. J. Macromol. Sci.,
Chem. , A25, 1667 (1988) .

Bog93 Bogoeva-Gaceva, G . ; Pimonenko, N.Y. and Petrov, G,
Text. Res. J., 63, 51 (1993).

Bor88 Borggreve, J.M. and Gaymans, R.J. Polymer, 29,
1441 (1988) .

Bow62 Bowen, R.L. U.S. Patent 3,066,112 (1962).

Bra8 9 Brandrup, J. and Immergut, E.H. Polymer Handbook,
Wiley & Sons, New York (1989) .

173

174

Bra86

Bra92

Bro89

Buc96a

Buc96b

Bye92

Cam85
Cam8 6
Che91
Chi96
Chu82
Chu84
Chu88a

Chu88b

Cor84
Cra80

Cra82

Brannstorm, M. J. Endod. , 12, 475 (1986).

Brady, R. J. M.S. -Rev. Macromol. Chem. Phys. C32,
135 (1992) .

Brown, H.R. Macromolecules, 22, 2859 (1989) .

Buchenska, J. J. Appl . Poly. Sci., 60, 519 (1996).

Buchenska, J. J. Appl. Poly. Sci., 61, 567 (1996).

Byerley, T.J.; Eick, J.D.; Chen, G.P.; Chappelow,
C.C. and Million, F. Dent. Mater., 8, 345 (1992).

Cameron, C.G. and Chisolm, M.S. Polymer, 26, 437
(1985) .

Cameron, C.G. and Chisolm, M.S. Polymer, 27, 1420
(1986) .

Chen, G.F. and Jones, F.N. Macromolecules, 24,
2151 (1991) .

Chiang, C. and Chang, F. J. Appl. Polym. Sci., 61,
2411 (1996) .

Chujo, Y.; Tatsuda, T. and Yamashita Y. Polym.
Bull. , 8, 239 (1982) .

Chujo, Y.; Kobayashi, H. and Yamashita, Y. Polym.
Commun. , 25, 278 (1984) .

Chujo, Y.; Hiraiwa, A.; Kobayashi, H. and
Yamashita, Y. J Polym. Sci., Polym. Chem Ed., 26,
2991 (1988) .

Chujo, Y.; Kobayashi, H. and Yamashita, Y. Polym.
J., 20, 407 (1988) .

Corner, T. Adv. Polym. Sci., 62, 95 (1984).

Craubner, H. J. Poly. Sci., Polym. Chem. Ed., 18,
2011 (1980) .

Craubner, H. J. Poly. Sci., Polym. Chem. Ed., 20,
1935 (1982) .

175

Cre89 Creton, C; Hui, H.C.Y.; Kramer, E.J.;

Hadziioannou, G. and Brown, H.R. Bull. Am. Phys .
Soc. , 34, 709 (1989) .

Dav91 Davidson, C.L.; Van Zeghbroeck, L. and Feilzer,
A.J. J. Dent. Res., 70, 880 (1991).

DeB73 De Boos, A.G. Polymer, 14, 587 (1973) .

Den89 Denizligil, S. And Yagci, Y. Polymer Bulletin, 22,
547 (1989) .

Ela92 Elangovan, V. J. and Saccubai, S. J. Appl . Polym.
Sci. , 44, 2179 (1992) .

Fay82 Fayt, R.; Jerome, R. and Teyssie, Ph. J. Poly.
Sci., Polym. Phys. Ed., 20, 2209 (1982).

Fri65 Friedman, M . ; Cavins, J.F. and Wall, J.S. J. Am.
Chem. Soc, 87, 3672 (1965).

Fur63 Furukawa, J.; Yamashita, S. And Okamoto, H. Japan
Patent 7146/63 (1963) .

Gai80 Gaillard, P.; Ossenbach-Sauter, M. and Reiss, G.
Makromol. Chem. Rapid Commun. , 1, 771 (1980).

Gna87 Gnanou, Y. and Rempp P. Makromol. Chem., 188, 2267

(1987) .

Gna88 Gnanou, Y. and Rempp P. Makromol. Chem., 189, 1997

(1988) .

God69 Godfrey, R.A. and Miller, G.W. J. Polym. Sci. Part
A-l, 7, 2387 (1969) .

Gon95a Gonzalez-Montiel, A.; Keskkula, H. and Paul, D.R.
Polymer, 36, 4587 (1995) .

Gon95b Gonzalez-Montiel, A.; Keskkula, H. and Paul, D.R.
Polymer, 36, 4605 (1995) .

Gon95c Gonzalez-Montiel, A.; Keskkula, H. and Paul, D.R.
Polymer, 36, 4621 (1995) .

Gra95 Graf, N. Master's Thesis, University of Florida
(1995) .

176

Ham84 Hamaide, T . ; Revillon, A. and Guyot, A. Euro.
Polym. J. , 20, 855 (1984) .

Has83 Hasimoto, T . ; Shibayama, M.; Fujimura, M. and
Kawai H. In: Block Copolymers Science and
Technology, Meier D.J., Ed., MM I Press, New York

(1983) .

Her74 Hergenrother, W.L. and Ambrose, R.J. J. Polym.
Sci., Polym. Chem. Ed., 12, 2613 (1974).

Hig80a Higashi, F.; Goto, M. and Kakinori, H. J. Polym.
Sci., Polym. Chem. Ed., 18, 1711 (1980).

Hig80b Higashi, F. and Taguchi, Y. J. Polym. Sci., Polym.
Chem. Ed. , 18, 1841 (1980) .

Hig80c Higashi, F.; Sano, K. and Kakinoki, H. J. Polym.
Sci., Polym. Chem. Ed., 18, 2875 (1980).

Imi84 Imanishi, Y. J. Macromol. Sci. Chem., A21, 1137

(1984) .

Ito77 Ito, K.; Hashizuka, H. and Yamashita, Y.
Macromolecules, 10, 821 (1977) .

Iza93 Izawa, M.; Nunomoto, S. and Kawakami, Y. Polym.
J. , 25, 873 (1993) .

Joh89 Johnston, A.D.; Asmussen, E. and Bowen, R.L. J.
Dent. Res., 68, 1337 (1989).

Kaw89 Kawaguchi, M.; Fukushima, T. and Horibe, T. Dent.
Mater. J, 8, 4 (1989) .

Kim92 Kimura, T . ; Tanji, K. and Minabe, M. Polym. J.,
24, 1311 (1992) .

Lee89 Lee, C.H. and Brauer, G.M. J. Dent. Res., 68, 484
(1989) .

Lee97 Lee, Y.M. and Shim, J.K. Polymer, 38, 1227 (1997)

Liu90 Liu, C.F.; Collard, S.M. and Armeniades, CD. Am.
J. Dent. , 3, 44 (1990) .

Maj92 Majumdar, B . ; Keskkula, H. and Paul, D.R. Proc.

ACS Div. Polym. Mater. Sci. Eng . , 67, 307 (1992).

177

Maj94a Majumdar, B . ; Keskkula, H. and Paul, D.R. J. Appl .
Polym. Sci., 54, 339 (1994).

Maj94b Majumdar, B.; Keskkula, H. and Paul, D.R. Polymer,
35, 1399 (1994) .

Mas79 Masar, B . ; Cefelin, P. and Sebenda, J. J. Poly.
Sci., Polym. Chem. Ed., 17, 2317 (1979).

Mas82 Masson, P.; Beinert, G.; Franta, E. and Rempp, P.
Polym. Bull., 1, 17 (1982).

Meh97 Mehlem, J.J.; Arnold, J. J.; Scwarz, M.W.; Zamora,
M.P. and Brennan, A.B. Am. Chem. Soc, Div. Polym.
Chem., Polym. Prepr . , 38, 159 (1997).

Mei73 Meier D.J. In: Block and Graft Copolymers, J.J.
Burke and V. Weiss, Eds., Syracuse University
Press, Syracuse, N.Y. (1973).

Mei90 Meijs, G.F. and Riddardo, E. J. Macromol. Sci.-
Macromol. Chem. Phys. C, 30, 305 (1990) .

Mis96 Misra, B.N.; Kaur, I.; Gupta, A.; John, V. and

Singha, A.S. Polym. & Polym. Comp . , 4, 411 (1996).

Mod93 Modic, M.J. and Pottick, L.A. Polym. Eng . Sci.,
33, 819 (1993) .

Mor96 Moraes, M.A.R.; Moreira, A.C.F.; Barbosa, R.V. and
Soares, B.G. Macromolecules, 29, 416 (1996) .

Mou93 Mougin, N . ; Rempp, P. and Gnanou, Y. J. Polym.
Sci., Polym. Chem. Ed., 31, 1253 (1993).

Mue93 Mueller-Schulte, D. Rad. Phys. Chem., 42, 891
(1993) .

Muh87 Muhlbach, K. and Percec, V. J. Polym. Sci., Polym.
Chem. Ed., 25, 2605 (1987).

Nai92 Nair, C.P.R. Euro. Polym. J., 28, 1527 (1992).

Nak80 Nakabayashi, N. and Masuhara, E. In: Biomedical

Polymers--Polymeric Materials and Pharmaceuticals
for Biomedical Use. E.P. Goldberg and A. Nakajuma,
eds., Academic Press, New York, 85-111 (1980).

178

Nat88 Natansohn, A.; Murali, R. and Eisenberg, A.

Makromol. Chem. , Macromol. Symp. , 16, 175 (1988).

Nos77 Noshay, A. and McGrath, J.E. Block Copolymers--

Overview and Critical Survey, Academic Press, New
York (1977) .

Osh92 Oshinski, A.; Keskkula, H. and Paul, D.R. Polymer,
33, 284 (1992) .

Ore97 Orefice, R. Ph.D. Dissertation, University of
Florida (1997) .

Pei86 Peiffer, D.G.; Duvdevani, I.; Agarwal, P.K. and
Lundberg, R.D. J. Polym. Sci. Lett. Ed., 24, 581
(1986) .

Pei94 Peiffer, D.G. and Rabeony, M . J. Appl . Polym.
Sci. , 51, 1283 (1994) .

Pet79

Peu91

Petit, D.; Jerome, R. and Teyssie, Ph. J. Polym.

Sci., Polym. Chem. Ed., 17, 2 903 (197 9).

Peutzfeldt, A. and Asmussen, E. J Dent Res., 70,
1537 (1991) .

Rem84a Rempp, P.F. and Franta, E. Adv. Polym. Sci., 58, 1
(1984) .

Rem84b Rempp, P.; Lutz, P.; Mason, P. and Franta E.
Makromol. Chem Suppl., 8, 3 (1984).

Ros82 Rosen, S.L. In: Fundamental Principles of

Polymeric Materials, Wiley & Sons, New York, 103-
128 (1982) .

Sch82 Schulz, G.O. and Milkovich, R. J. Appl. Polym.
Sci. , 27, 4773 (1982) .

Sch84a Schulz, G.O. and Milkovich R. J. Appl. Polym. Sci.
Chem. Ed., 22, 3795 (1984).

Sch84b Schulz, G.O. and Milkovich R. J. Appl. Polym. Sci.
Chem. Ed., 22, 1633 (1984).

Sch97 Schwarz, M.W.; Zamora, M.P.; Mehlem, J.J. and

Brennan, A.B. Am. Chem. Soc, Div. Polym. Chem.,
Polym. Prepr., 38, 161 (1997).

179

Sea95 Seadan, M . ; Lambla, M.; Narkis, M.; Siegmann, A.
and Tzur, A. Polym. Adv. Tech., 6, 679 (1995).

Shu90 Shull, K.R.; Kramer, E.J.; Hadziioannou, G. and
Tang, W. Macromolecules , 23, 4780 (1990) .

Sil91 Silverstein, R.M.; Bassler, G.C. and Morrill T.C.
Spectrometric Identification of Organic Compounds,
Wiley & Sons, New York (1991) .

Shu91 Shull, K.R.; Winey, K.I.; Thomas, E.L. and Kramer,
E.J. Macromolecules, 24, 2748 (1991).

Sod91 Soderholm K.J.M. Dent. Mater., 7, 1 (1991).

Sta91 Stansbury, J.W. Macromolecules, 24, 2029 (1991) .

Ste77 Stehlicek, J. and Sebenda, J. Eur. Polym. J. 13,
949 (1977) .

Ste82 Stehlicek, J. And Sebenda, J. Eur. Polym. J., 18,
535, (1982) .

Sul93 Suliman, A. A.; Boyer, D.B. and Lakes, R.S. J.
Dent. Res., 72, 1532 (1993).

Tan90 Tang, S . C . ; Hu, C.P. and Ying, S.K. Polym. J., 22,
70 (1990) .

Tsu91 Tsukahara, Y . ; Nakanishi, Y.; Yamashita, Y.;

Ohtani, H.; Nakashima, Y . ; Luo, Y.F.; Ando, T. and
Tsuge, S. Macromolecules, 24, 2493, (1991) .

Ven93 Venz, S. and Dickens, B. J. Dent. Res., 72, 582
(1993) .

Voe90 Voet, D. and Voet, J.G. In: Biochemistry, Wiley &
Sons, New York, 62, 1990.

Won82 Wondraczek, R.H. and Kennedy, J. P. J. Polym. Sci.,
Polym. Chem. Ed., 20, 173 (1982).

Wes97 West, J.K.; Brennan, A.B.; Clark, A.E.; Zamora, M.
and Hench, L.L., accepted by J. Biomed. Mater.
Res. (1997) .

WuC93 Wu, C.J.; Kuo, J.F. and Chen C.Y. Polym. Eng .
Sci. , 33, 1329 (1993) .

180

Yam81 Yamashita, Y.; Chujo, Y.; Koyabayashi, H. and
Kawakami, H. Polym. Bull., 5, 361 (1981).

YnM94 Yn, M.S. and Ma, CM. J. Appl . Polym. Sci., 53,
213 (1994) .

Yos94 Yoshimitsu, Y . ; Sugiyama, K.; Shiraishi, S . ; Oda,
T. and Ohashi, T. Poly. J., 26, 551 (1994).

You95 Young, M.L.; Sung, Y.I.; Jin, K.S.; Jin, H.K.;

Chong, S.C. and Yong, K.S. Polymer, 36, 81 (1995)

Zam95 Zamora, M.P. and Brennan, A.B. Yearly Progress
Report for Dow Chemical Co., University of
Florida, Gainesville, Florida (1995) .

Zha90 Zhang, X. and Eisenberg, A. Polym. Adv. Tech., 1,
9 (1990) .

BIOGRAPHICAL SKETCH

In keeping with his current sleeping schedule, Michael
Perez Zamora was born at approximately 3:00 A.M. on February
28, 1970, in Cincinnati, Ohio, to Pablo and Sylvia Zamora
and to Bella Acosta, his grandmother. Approximately nine
years earlier, Pablo and Sylvia emigrated from Cuba and met
in Miami, Florida. They moved to Gainesville, Florida and
married during the time Pablo earned his master's degree in
chemical engineering at the University of Florida, almost
thirty years ago.

Michael lived in Cincinnati for three years before his
father was transferred to Venezuela. After four years in
Venezuela, and with Cuban Spanish speaking parents, he came
back to Cincinnati a purely Latin kid who spoke little
English. Michael adapted quickly to the American way of
life upon exposure to Sesame Street, rollercoasters, theme
parks, and baseball. Michael lived in Cincinnati until his
family moved to Longmeadow, Massachusetts in 1984. During
his time in high school, Michael was heavily involved in
extracurricular activities including being named captain of
the varsity swim and tennis teams.

After graduating from high school in 1987, Michael
enrolled at Georgetown University in Washington, D.C. It

181

182

was during his time at Georgetown that he starting dating a
longtime friend from high school days, the girl he would
eventually marry, Karen Morey. After four years of having
entirely too much fun, Michael went to work as an intern at
DuPont, where he was first exposed to the real life
implications of what he was learning at school. Michael
graduated from Georgetown in the Fall of 1991 with a B.S. in
chemistry and returned to DuPont for another internship of
four months .

After taking the summer off to sharpen up his golf game
in preparation for graduate school, Michael began his
graduate studies under the supervision of Dr. Anthony
Brennan at the University of Florida in Gainesville in
August 1992. Dr. Brennan, a professor of only one year at
the time, quickly introduced Michael to the harsh and
important realities of life. Michael settled in quickly
with the help of Karen, whom he married at the end of his
second year of graduate work.

Life has come full circle for Michael. As he prepares
to embark on his new career with Paxon, a division of Exxon
Chemical Co. in Baton Rouge, Louisiana, he and Karen are
leaving Gainesville almost exactly thirty years after his
newly married parents, Pablo and Sylvia, left in their
pursuit of the American dream.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Debtor of Philosophy.

JA^didn.

Anthony Jbt Br ennan, Chair
Associate Professor of

Materials Science and

Engineering

-7

fames H. Adair
'Associate Professor of
Materials Science and
Engineering

Christopher D. Batich
Professor of Materials
Science and Engineering

Eugene PV Goldberg
Professor of Materials
Science and Engineering

fCiw^lki \j l^>i^ l

Kenneth B. Wagener /
Professor of Chemistry

This dissertation was submitted to the Graduate Faculty
of College of Engineering and to the Graduate School and was
accepted as partial fulfillment of the^requirements for the
degree of Doctor of Philosophy.

December, 1997

:red M. Phillips
Dean, College of Engineering

Karen A. Holbrook
Dean, Graduate School

1780
1997

.125

UNIVERSITY OF FLORIDA

3 1262 08555 1074